Lyme neuroborreliosis is a tick-borne infectious disease caused by the gram-negative spirochete bacterium
Borrelia burgdorferi sensu lato. Transmission of
B. burgdorferi occurs via bites by the tick species Ixodes ricinus in Europe and Ixodes scapularis and Ixodes pacificus in the United States. Different species of
B. burgdorferi show distinct patterns of distribution, whereas
Borrelia garinii, B. afzelii, and
B. burgdorferi sensu stricto are common in Europe; in the USA,
B. burgdorferi sensu stricto predominates [
1]. The different genospecies seem to be associated with distinct clinical manifestations, i.e.,
B. garinii being more closely associated with neurological manifestations of borreliosis than other species [
1]. Affection of the nervous system occurs in approximately 15% of all patients with lyme borreliosis [
2]. Common clinical manifestations of neuroborreliosis are polyradiculoneuritis (Bannwarth’s syndrome, 63.75% of all cases), in 60% accompanied by cranial nerve dysfunctions (most frequently facial palsy), and meningitis (23.75%) [
3,
4]. To a lesser extent, affections of the central nervous system like encephalitis and myelitis occur (12.5%) [
4]. Other, rarer manifestations are borrelia-induced vasculitis (0.3%) and myositis (case reports) [
5‐
8]. Beside nervous system affections,
B. burgdorferi sensu lato can affect multiple other organ systems. In Europe, manifestations include the early developing erythema migrans (89%), acrodermatitis atrophicans (1%), lyme arthritis (5%), or lyme carditis (<1%) [
1,
9]. Diagnosis of neuroborreliosis is usually based on clinical presentation, serologic testing, and analysis of cerebrospinal fluid (CSF) [
10]. Tiered case definitions exist regarding likelihood of diagnosis depending on diagnostic results [
3,
11]. Different authors report variable sensitivities of this approach, and other diagnostic strategies of unclear accuracy have been suggested [
12‐
14]. Controversy exists also on the field of therapy, where choice, route of administration, and length of treatment are subject of intense debate. The guidelines of the Infectious Diseases Society of America (IDSA), the European Federation of Neurological Societies (EFNS), and the evidence-based practice parameters of the American Academy of Neurology (AAN) recommend antibiotic treatment with a duration of up to 21–28 days [
10,
15,
16], whereas guidelines of the International Lyme and Associated Diseases Society (ILADS) states that several months of antibiotic therapy are often required [
17]. Furthermore, the ILADS guideline states that short-course antibiotic treatment shows high failure rates for treatment of neuroborreliosis. Choice of antibiotic agent for treatment of neuroborreliosis is also a matter of ongoing controversy. IDSA, EFNS, and AAN guidelines usually recommend cephalosporin antibiotics, doxycycline, or penicillin antibiotics, whereas some authors recommend treatment with other substances, like carbapenem antibiotics, metronidazole, or antimalarial drugs such as hydroxychloroquine for certain subgroups of patients [
17]. The use of combination therapy (use of multiple antibiotics concurrently) for the treatment of neuroborreliosis is advocated by the ILADS, whereas the IDSA, EFNS, and AAN discourage the use of antibiotic combination regimes. The optimal dosage of the orally administered antibiotic doxycycline remains unclear. Studies investigating the effects of doxycycline on people with neuroborreliosis examined daily doses ranging from 200–400 mg [
18‐
20]. Accordingly, the IDSA and EFNS guidelines recommend doxycycline doses of 200 mg, whereas the guidelines of the Deutsche Gesellschaft für Neurologie (DGN) and AAN guidelines suggest applying higher doses of up to 400 mg [
10,
15,
16,
21]. These various, partly contradicting recommendations lead to a considerable ambiguity and doubt in patients and healthcare providers alike when facing treatment decisions for neuroborreliosis. Therefore, it seems necessary to review, evaluate, and summarize the available evidence for drug treatment of neuroborreliosis for making evidence-based clinical recommendations. To adequately consider the wealth of research that has been conducted and to overcome limitations of availability of only few RCTs for a limited number of important questions with regard to optimal treatment, this review will evaluate randomized controlled trials and non-randomized studies for treatment of neuroborreliosis. Adding non-randomized studies to systematic reviews can have certain advantages. Non-randomized studies seem to be more suitable for the detection of adverse events than RCTs and may have longer follow-up periods [
22].