Lyme borreliosis (LB) is the most common tick-borne disease in the northern hemisphere [1]. In the Netherlands, a fourfold increase was observed in the number of consultations for Lyme borreliosis in primary care between 1994 and 2018, with an increase in the estimated incidence from 39 to 148 per 100,000 inhabitants, respectively [2,3,4]. The disease is caused by several Borrelia species within the Borrelia burgdorferi sensu lato complex, with the predominant species in Europe being B. garinii, B. afzelii, and B. burgdorferi sensu stricto, whereas in North America, it is mainly caused by B. burgdoferi ss [5,6]. In humans, LB primarily affects the skin, nervous system, heart, or joints and typically manifests as a skin lesion with an expanding bull’s-eye pattern called erythema migrans (EM), which appears several days to weeks after a tick bite. The skin lesion may be accompanied by flu-like symptoms such as fever, myalgia, and headache. Later manifestations of the disease, such as neuroborreliosis, acrodermatitis chronica atrophicans (ACA), or arthritis, can present weeks or months after infection [7,8,9,10,11,12]. Unfortunately, even after the recommended antibiotic treatment, on average, 5–10% of patients report symptoms for years after infection [13], after which neuroborreliosis can reach up to 48% [14,15]. This “chronic” form of borreliosis was recently described by Steere [16]. Considering Borrelia diagnostics, post-treatment Lyme disease syndrome (PTLDS) and central sensitization syndrome are the most challenging entities [12].

The symptoms in PSL are often non-specific and vary from chronic fatigue and musculoskeletal pain to neurocognitive difficulties, and the criteria were previously described by Wormser [17]. Although chemokine ligand 13 (CXCL13) is a recognized biomarker for acute neuroborreliosis [18], there are no specific biological markers for PSL, making it extremely difficult to define the disease. Furthermore, the etiology is unclear, and possible explanations range from persistent infection with Borrelia spp. to factors not related to LB [16].

Standard serological diagnosis relies on a two-tier approach, starting with a Borrelia-specific IgG and IgM enzyme-linked immunosorbent assay (ELISA). If the screening yields a positive or equivocal result, confirmation follows by testing the IgG and IgM reactivity by Western blotting (WB) against several Borrelia-specific proteins [19]. A modified two-tier approach has recently been evaluated and proven equally sensitive and specific. This modified approach consists of two consecutive Borrelia-specific ELISAs from different commercial brands [20,21,22].

Serological diagnostics are known to have poor positive predictive value in populations with a low pretest probability. Stage and symptom entities influence pretest probabilities and the level of false-positive serological results [3,23,24]. IgM reactivity is especially hampered by a high proportion of false-positive reactions [25]. Therefore, Borrelia serology should only be performed if a high pre-test probability is present. Further, laboratory specialists should only consider the serological results in combination with clinical information. Finally, clinicians should take great care not to trust alternative testing, as this might lead to misdiagnosing LB [26,27].

Early borreliosis includes stage 1 and stage 2 diseases. Acute borreliosis (stage 1 disease) is often defined as symptoms developing within 30 days after exposure; however, the exact date of exposure is often difficult to pinpoint, and some scientists also define symptoms that develop within 10 weeks after exposure as early borreliosis. A large difference in the time span will influence the sensitivity of the serological test. Mounting a Borrelia-specific IgM response typically takes 2–4 weeks, and developing a Borrelia-specific IgG response may take up to two months [28,29]. Thus, the so-called “window period” in which no antibodies are detectable varies from several days to 2 months after the initial infection. When evaluating the performance of a serological test in the case of borreliosis, it is better to relate the findings to a clinical entity. Early borreliosis is best mirrored by the presence of EM, also referred to as early localized disease. In EM, the two-tier serological approach has been shown to have a limited sensitivity of 31–50% [3,29,30]. The early response is predominantly an IgM reaction with reactivity against outer surface protein C (OspC; 25 kDa), basic membrane protein A (BmpA; 39 kDa), and flagellin protein (41 kDa). Only 22% of serum samples in early localized disease have been shown to be reactive in a combination of IgG ELISA and WB, with a predominant protein reactivity against surface lipoprotein E (VlsE; 35 kDa) [31]. The intensity of the antibody response is related to the duration of the EM before antibiotic treatment and the extent of dissemination [31,32]. Because of the limited sensitivity of serology in early localized disease, it is advised to treat patients directly when EM is diagnosed and not rely on serological diagnostics [33].

In early disseminated disease (stage 2, acute neuroborreliosis, lymphocytoma, and carditis), the two-tier serology approach has a sensitivity of 63–77% [3,28,30]. In comparison to patients with EM, patients with neuroborreliosis have lower IgM optical density (OD) values and higher IgG OD values against B. burgdorferi antigens [34]. Cerebrospinal pleiocytosis, intrathecal antibody production of IgM and IgG isotypes, and an increased level of CXCL-13 in the cerebrospinal fluid are typically present in patients with neuroborreliosis [35].

Although IgM testing has proven to be sensitive in early Borrelia diagnostics, a high proportion of false-positive results has also been seen, especially if the IgM reactivity exceeded 1 month after exposure, with no evidence of IgG seroconversion [25,36]. Branda and colleagues [30] suggested an alternative two-tier approach using the IgG ELISA reactivity confirmed by one WB VlsE band. This approach reached 96% sensitivity in early disseminated disease and is, therefore, an adequate alternative to IgM testing at this disease stage and onward.

Late disseminated borreliosis (stage 3) presents as ACA, Lyme arthritis, and/or continuing neurological disease activity lasting more than 6 months [12]. The two-tier approach, including 5 of 11 IgG-positive WB bands, showed 95–100% sensitivity in this patient group [30]. To enhance specificity, a positive IgM immunoblot alone, in the absence of a positive IgG immunoblot, should not be interpreted as active LB in patients with an illness duration of more than 30 days [28].

The clinical manifestations, diagnostics, and antibiotic treatment of stages 1, 2, and 3 borreliosis are well-established. Most patients will recover, and the Borrelia-related inflammation will normalize. However, a small proportion of patients continue to suffer from non-specific symptoms consisting of fatigue, arthralgia, myalgia, and perceived cognitive impairment. These symptoms will typically persist in spite of antibiotic treatment. Case definitions show large variations, and diagnostic workup is difficult and often accompanied by confusion and controversy.