Cerebrospinal Fluid Profiles Across Clinical Manifestations of Early European Lyme Neuroborreliosis Authors: Katarina Ogrinc, Petra Bogovič, Andrej Kastrin, Eva Ružić-Sabljić, Franc Strle Published: 2026 ABSTRACT AbstractBackgroundEarly European lyme neuroborreliosis (LNB) presents with distinct clinical manifestations, but comparative data on cerebrospinal fluid (CSF) findings across these forms are limited.MethodsWe conducted a retrospective study of 405 patients with confirmed early LNB treated at a single center between 2005 and 2025. Thirteen routine CSF parameters were analyzed across three main clinical manifestations—meningoradiculoneuritis (Bannwarth syndrome), cranial neuritis, and erythema migrans (EM)-associated meningitis—and in relation to the duration of neurological symptoms. Multivariate regression was used to assess independent associations.ResultsOf 405 patients, 274 (68%) had Bannwarth syndrome, 82 (20%) cranial neuritis, and 49 (12%) EM-meningitis. CSF abnormalities differed significantly among groups and were most pronounced in Bannwarth syndrome and least marked in EM-meningitis. Patients with Bannwarth syndrome more frequently had positive borrelial serology and intrathecal antibody synthesis. CSF leukocyte counts, protein, and albumin increased during the first 4 weeks of disease and then stabilized, whereas immunoglobulin levels peaked around week 6. Longer duration of neurological symptoms was associated with higher lymphocyte proportion, increased protein levels, and elevated IgM and IgG concentrations. Multivariate analysis confirmed that clinical manifestation, symptom duration, age, and sex independently influenced CSF parameters.ConclusionsThe three principal manifestations of early European LNB differ in the magnitude of CSF inflammation and blood–brain barrier dysfunction. Bannwarth syndrome is associated with the most pronounced intrathecal inflammatory response, whereas EM-associated meningitis represents a milder phenotype. CSF abnormalities evolve over time, underscoring the importance of early diagnostic evaluation.Keywords: Bannwarth syndrome, borrelial meningitis, cerebrospinal fluid, cranial neuritis, lyme neuroborreliosis Lyme borreliosis (LB) is the most common tick-transmitted infection in the Northern Hemisphere. It is caused by Borrelia burgdorferi sensu lato (s. l.) [1]. The earliest and most frequent manifestation is erythema migrans (EM), which develops days to weeks after a tick bite [2]. If untreated, the infection may disseminate, resulting in multiple EM skin lesions and/or involvement of other organ systems, including the nervous system, joints, and heart [3, 4]. In Europe, lyme neuroborreliosis (LNB) represents the second most frequent manifestation of disseminated LB after multiple EM and is predominantly associated with Borrelia garinii infection [5–8]. Early European LNB most commonly presents as: (1) meningoradiculoneuritis (Bannwarth syndrome), characterized by severe radicular pain, sometimes accompanied by cranial nerve involvement; (2) cranial neuritis without radicular pain, most frequently unilateral peripheral facial nerve palsy; and (3) lymphocytic meningitis [3, 4, 9, 10]. Although the clinical characteristics and outcomes of these manifestations have been extensively described, systematic comparative data on cerebrospinal fluid (CSF) findings across different clinical forms of early LNB remain limited. In addition, the temporal evolution of routine CSF parameters in relation to the duration of neurological symptoms is insufficiently characterized. To our knowledge, no previous study has systematically compared routine CSF findings across the principal clinical phenotypes of early European LNB in a large, well-characterized cohort. The aim of this present study was to evaluate and compare routine CSF findings in patients with borrelial meningoradiculoneuritis, cranial neuritis and meningitis, and to assess their relationship with the duration of neurological involvement. PATIENTS AND METHODS Ethics This retrospective study was approved by the Medical Ethics Committee of the Ministry of Health of the Republic of Slovenia (0120-552/2023/3, 0120-656/2025-2711-3). Lumbar puncture and skin biopsy were performed after obtaining written informed consent. Patients The study was conducted at the LB Outpatient Clinic, Department of Infectious Diseases, UMC Ljubljana. Patients aged ≥15 years with confirmed early LNB diagnosed between 2005 and 2025 were eligible. All patients were tested for tick-borne encephalitis virus (TBEV) infection; those with evidence of recent TBEV infection were excluded. Thirteen CSF laboratory parameters obtained as a part of routine clinical diagnostics were evaluated in every patient and compared: (1) across the three principal clinical manifestations of LNB—Bannwarth syndrome, cranial neuritis and EM-associated meningitis; and (2) according to the duration of neurological symptoms and signs prior to CSF examination. The clinical presentation of part of this cohort has been reported previously [6, 11–14]. Definitions Lyme Neuroborreliosis All patients fulfilled modified EFNS criteria for definite LNB. Specifically, the first two criteria are identical to the EFNS criteria [9], whereas the third criterion additionally accepts isolation of Borrelia from CSF and/or concomitant EM as microbiological evidence of infection. Accordingly, all patients fulfilled the following three criteria: (1) clinical symptoms/signs compatible with LNB, (2) CSF pleocytosis (>5×10⁶ leukocytes/L), and (3) intrathecal borrelial IgM and/or IgG antibody synthesis (ITBAS) and/or isolation of Borrelia from CSF and/or concomitant EM. Clinical Manifestations Borrelial meningoradiculoneuritis (Bannwarth syndrome): patients who presented with new-onset radicular pain; Borrelial meningoradiculoneuritis (Bannwarth syndrome): patients who presented with new-onset radicular pain; Cranial neuritis: cranial nerve involvement without radicular pain; Cranial neuritis: cranial nerve involvement without radicular pain; EM-associated meningitis (hereinafter “EM-meningitis”): EM skin lesion(s) with symptoms suggestive of central nervous system involvement, without cranial nerve palsy or radicular pain. EM-associated meningitis (hereinafter “EM-meningitis”): EM skin lesion(s) with symptoms suggestive of central nervous system involvement, without cranial nerve palsy or radicular pain. When patients fulfilled criteria for more than one clinical manifestation, they were classified according to the following predefined hierarchy: meningoradiculoneuritis, cranial neuritis, and EM-associated meningitis. Duration of Neurologic Symptoms/Signs Defined as the number of days from onset of neurologic manifestations to CSF examination. CSF Parameters CSF was analyzed for cell counts and protein concentration. Pleocytosis was defined as >5 × 106 leukocytes/L, and elevated protein concentration as >0.45 g/L. Differential leukocyte counts (lymphocytes, monocytes, neutrophils) were determined, and lymphocyte proportion was calculated. CSF glucose concentration and the CSF/blood glucose ratio were assessed. Albumin and immunoglobulin M (IgM) and G (IgG) concentrations were measured in serum and CSF, and corresponding quotients were calculated. Elevated quotients were interpreted as indicative of blood–brain barrier dysfunction. Borrelia Burgdorferi Sensu Lato Serology Up to 2010, antibodies were determined using an indirect immunofluorescence assay (IFA) without absorption [15, 16]. From 2011 onwards, a chemiluminescence immunoassay (LIAISON®, Diasorin, Italy) was used [17]; results were interpreted according to the manufacturer's instructions. ITBAS was assessed using the Reiber and Peter method [18]. Antibody index values >2.0 (IFA) or ≥1.5 (LIAISON®) indicated ITBAS. TBEV Serologic Testing All patients were tested for serum IgM and IgG antibodies to TBEV. Enzygnost Anti-TBE/FSME Virus (IgM, IgG) test (Siemens AG, Munich, Germany) was used from 2005 up to September 2023, according to the manufacturer's instructions. Since then, Anti-TBE Virus ELISA IgG/IgM Kit (Euroimmun, Lübeck, Germany) has been used. The presence of specific IgM and IgG antibodies in serum was interpreted as evidence of a recent infection with the TBEV. Cultivation of B. Burgdorferi Sensu Lato Cultivation from CSF, blood and skin specimens was performed using modified Kelly–Pettenkofer (MKP) medium, as previously described [19, 20]. Blood (9 ml) and CSF (1 ml) samples were obtained from most patients. In patients with EM who consented to skin biopsy, a 3 mm punch biopsy was obtained either from the margin of an existing EM lesion or from the site of a previous lesion. CSF and skin samples were inoculated directly into tubes containing 7 ml MKP medium. Blood samples were centrifuged and 1 ml of plasma was inoculated into MKP medium. Cultures were incubated at 33°C and examined weekly by dark-field microscopy for up to 9 weeks (CSF and skin) and 12 weeks (blood) [20]. Isolates were identified to species/strain level by pulsed-field gel electrophoresis following MluI restriction of genomic DNA or by PCR-based restriction fragment length polymorphism analysis of the 5S-23S intergenic region [20, 21]. Statistical Methods Descriptive statistics were used to summarize the dataset, with continuous variables presented as medians and interquartile ranges (IQRs) and categorical variables reported as frequencies and percentages with 95% confidence intervals (CIs). To evaluate differences between groups, we employed Student's t-tests or Mann–Whitney U tests for continuous variables, depending on the normality of the distribution, while chi-square tests or Fisher's exact tests were used for categorical data as appropriate. Relationships between variables were further analyzed using linear regression for continuous outcomes, binomial logistic regression for binary responses, and multinomial logistic regression for multicategorical variables. Associations are reported as either linear regression coefficients (β) or odds ratios (ORs), alongside 95% CIs and P values. To mitigate the risk of Type I errors stemming from multiple comparisons, we applied the Benjamini-Hochberg procedure. All analyses were conducted using R Statistical Software (The R Project for Statistical Computing, https://www.r-project.org). RESULTS In the period 2005–2025, 405 patients with early LNB were included (232 males, 173 females, median age 56 years). Of these, 274 (68%) had meningoradiculoneuritis (Bannwarth syndrome), 82 (20%) cranial neuritis, and 49 (12%) EM-meningitis. Patients with Bannwarth syndrome were significantly older and had longer duration of symptoms prior to lumbar puncture compared with the other groups (P < .001) (Table 1). Table 1.Demographic and Clinical Characteristics of Patients With Early Lyme Neuroborreliosis by Clinical ManifestationVariableBannwarth Syndrome (n = 274)Cranial Neuritis (n = 82)EM-associated Meningitis (n = 49) P ValueMale sex, n (%)150 (55)55 (67)27 (55).130Age, years61 (52–70)47 (35–60)48 (36–59)<.001Duration of neurologic involvement, days24 (14–35)10 (6–18)10 (5–17)<.001Data are presented as median (IQR) or number (%).EM, erythema migrans. Demographic and Clinical Characteristics of Patients With Early Lyme Neuroborreliosis by Clinical Manifestation Data are presented as median (IQR) or number (%). EM, erythema migrans. CSF Findings by Sex, Age, and Prior Antibiotic Treatment No significant sex-related differences in CSF parameters were observed. Older patients had lower leukocyte counts but higher glucose levels and a higher proportion of elevated protein concentrations (Supplementary Table 1). Ninety-five patients (23.5%) had received antibiotic therapy active against Borrelia prior to lumbar puncture. Neither CSF parameters nor the duration of neurologic involvement differed significantly between previously treated and untreated patients (Supplementary Table 2). CSF Findings by Clinical Manifestation Significant differences in CSF parameters were observed among clinical groups, including leukocyte count, total protein, albumin, IgM and IgG concentrations, as well as albumin and immunoglobulin quotients. Abnormalities were most pronounced in Bannwarth syndrome and least in EM-meningitis (Table 2; Figure 1). Table 2.Cerebrospinal Fluid Findings in Patients With Three Clinical Manifestations of Early European Lyme NeuroborreliosisVariableAll (N = 405)Bannwarth Syndrome (n = 274)Cranial Neuritis (n = 82)EM-associated Meningitis (n = 49) P ValueBS versus CNBS versus EM-MCN versus EM-MLeukocytes, ×106/L121 (48–293)157 (70–310)66 (24–266)37 (10–166).001<.001.130Lymphocytes, ×106/L107 (37–256)139 (59–267)59 (22–240)34 (8–155).001<.001.132Neutrophils, ×106/L4 (0–8)5 (0–9)3 (0–11)1 (0–5).997.025.139Lymphocyte/leukocyte ratio0.89 (0.81–0.94)0.89 (0.81–0.93)0.90 (0.81–0.95)0.88 (0.78–0.94).493.920.607Protein, g/L1.12 (0.61–1.93)1.40 (0.78–2.24)0.77 (0.51–1.38)0.64 (0.50–0.88)<.001<.001.136Glucose, mmol/L2.9 (2.6–3.3)2.9 (2.6–3.4)3.0 (2.6–3.2)3.0 (2.6–3.2).494.920.687CSF/serum glucose ratio0.54 (0.47–0.62)0.54 (0.47–0.61)0.55 (0.49–0.61)0.56 (0.50–0.62).489.233.687Albumin, mg/L623 (361–1030)769 (459–1195)455 (316–774)385 (294–559)<.001<.001.136IgM, mg/L9.33 (1.90–30.60)14.50 (3.53–37.80)4.95 (1.65–26.60)1.06 (0.46–7.29).004<.001.007IgG, mg/L99.6 (50.0–217.0)153.5 (66.2–261.0)66.4 (40.3–133.0)46.3 (31.0–74.3)<.001<.001.030Albumin quotient, ×10−314.4 (8.2–24.5)17.3 (10.4–27.4)10.4 (7.4–18.2)8.9 (6.4–11.6)<.001<.001.130IgM quotient, ×10−38.5 (1.6–30.9)13.2 (3.2–39.5)4.0 (1.1–20.1)1.2 (0.5–6.8).001<.001.026IgG quotient, ×10−310.0 (4.2–20.8)14.0 (6.1–23.8)6.4 (3.6–13.0)4.2 (3.3–6.6)<.001<.001.036BS, Bannwarth syndrome; CN, cranial neuritis; CSF, cerebrospinal fluid; EM, erythema migrans; EM-M, erythema migrans –associated meningitis.Data are presented as median (IQR). P values were adjusted for multiple testing.Leukocyte counts were corrected in traumatic lumbar puncture.Quotients represent CSF/serum ratios. Cerebrospinal Fluid Findings in Patients With Three Clinical Manifestations of Early European Lyme Neuroborreliosis BS, Bannwarth syndrome; CN, cranial neuritis; CSF, cerebrospinal fluid; EM, erythema migrans; EM-M, erythema migrans –associated meningitis. Data are presented as median (IQR). P values were adjusted for multiple testing. Leukocyte counts were corrected in traumatic lumbar puncture. Quotients represent CSF/serum ratios. Comparison of cerebrospinal fluid parameters across three clinical manifestations of early Lyme neuroborreliosis. Box-and-whisker plots showing distribution of selected CSF parameters in patients with Bannwarth syndrome (BS), cranial neuritis (CN), and EM-meningitis (M). Boxes represent interquartile range, horizontal lines indicate medians, and whiskers represent range. Borrelial serologic responses followed a similar gradient, with highest positivity rates and antibody levels in Bannwarth syndrome. The overall positivity rates for borrelial culture from CSF, blood and skin were 10%, 2% and 28%, respectively, without significant differences between clinical groups (Table 3). Additional data on serum and CSF antibody findings in patients with positive CSF cultures, as well as the small number of patients with CSF antibody positivity despite negative serum serology, are presented in the revised Table 3. Table 3.Borrelial Serologic and Culture Results in Patients With Three Clinical Manifestations of Early European Lyme NeuroborreliosisVariableAll (N = 405)Bannwarth Syndrome (n = 274)Cranial Neuritis (n = 82)EM-associated Meningitis (n = 49) P ValueBS versus CNBS versus EM-MCN versus EM-MBorrelial serologyPositive serum IgM204 (50.4%; 45.4–55.4)140 (51.1%; 45.0–57.2)38 (46.3%; 35.3–57.7)26 (53.1%; 38.3–67.5).794.877.794Positive serum IgG350 (86.4%; 82.7–89.6)255 (93.1%; 89.4–95.8)65 (79.3%; 68.9–87.4)30 (61.2%; 46.2–74.8).001<.001.042Positive CSF IgM269a (66.4%; 61.6–71.0)201 (73.4%; 67.7–78.5)48 (58.5%; 47.1–69.3)20 (40.8%; 27.0–55.8).020<.001.070Positive CSF IgG339a (83.7%; 79.7–87.2)249 (90.9%; 86.8–94.0)66 (80.5%; 70.3–88.4)24 (49.0%; 34.4–63.7).017<.001.001Positive borrelial IgM antibody index205/404 (50.7%; 45.8–55.7)155 (56.6%; 50.5–62.5)38 (46.3%; 35.3–57.7)12/48 (25.0%; 13.6–39.6).129<.001.037Positive borrelial IgG antibody index311/403 (77.2%; 72.8–81.2)231 (84.3%; 79.5–88.4)58/81 (71.6%; 60.5–81.1)22/48 (45.8%; 31.4–60.8).014<.001.007Borrelial antibody levels (U/mL), all tested patientsSerum IgM27.4 (10.5–79.7), n = 34124.3 (10.0–55.5), n = 24839.3 (11.5–118.0), n = 6459.7 (14.7–190.0), n = 29.016<.001.027Serum IgG229.1 (83.2–662.1), n = 338240.5 (117.7–810.7), n = 248211.1 (43.3–444.2), n = 6164.7 (15.8–168.2), n = 29.002<.001.112Borrelial antibody levels (U/mL), only positive resultsSerum IgM63.0 (36.3–158.6), n = 19450.2 (34.1–102.1), n = 136104.9 (66.7–190.0), n = 38117.7 (50.1–480.5), n = 20.705.705.705Serum IgG239.4 (114.5–721.1), n = 315241.0 (125.3–813.8), n = 241231.1 (101.7–503.7), n = 52108.0 (42.5–218.4), n = 22<.001<.001.045CSF borrelial antibody levels (U/mL), all tested patientsCSF IgM19.4 (3.3–73.8), n = 33423.0 (4.4–85.5), n = 24316.3 (2.4–51.7), n = 634.1 (0.2–40.2), n = 28.025<.001.028CSF IgG240.0 (45.6–1725.0), n = 333284.4 (84.7–2110.0), n = 243150.8 (11.5–720.5), n = 624.6 (0.1–161.4), n = 28.015<.001.047CSF borrelial antibody levels (U/mL), only positive resultsCSF IgM36.7 (14.9–121.1), n = 24938.0 (15.0–132.1), n = 18928.1 (13.3–104.9), n = 4640.2 (20.1–190.0), n = 14.046<.001.005CSF IgG282.3 (106.3–2048.0), n = 294430.8 (123.3–2292.0), n = 230224.8 (26.6–1270.0), n = 51173.7 (46.3–1239.0), n = 13<.001<.001<.001Borrelial culturePositive CSF culture36b/368 (9.8%; 7.0–13.3)23c/246 (9.3%; 6.0–13.7)10d/75 (13.3%; 6.6–23.2)3e/47 (6.4%; 1.3–17.5).554.779.554Positive blood culture5/338 (1.5%; 0.5–3.4)1f/227 (0.4%; 0–2.4)2g/69 (2.9%; 0.4–10.1)2h/42 (4.8%; 0.6–16.2).201.184.630Positive skin culture37/127 (28.3%; 20.7–37.0)28i/89 (31.5%; 22.0–42.2)3j/12 (25.0%; 5.5–57.2)6k/26 (19.2%; 6.6–39.4).118.619.118Positive skin culture in previously untreated patients36/106 (33.0%; 24.2–42.8)27/75 (36.0%; 25.2–47.9)3/9 (33.3%; 7.5–70.1)6/22 (22.7%; 7.8–45.4).165.610.165BS, Bannwarth syndrome; CN, cranial neuritis; CSF, cerebrospinal fluid; EM, erythema migrans; EM-M, erythema migrans-associated meningitis.Data are presented as number (%; 95% CI) or median (IQR).Antibody concentrations were available only in the subgroup tested by chemiluminescence immunoassay (LIAISON). aTwo patients with positive IgM and/or IgG CSF antibodies were seronegative. bAmong 36 patients with positive CSF culture, 28 (78%) had positive serum IgM and/or IgG borrelial antibodies; 26 (72%) had positive CSF IgM and/or IgG borrelial antibodies; 21 (58%) had positive IgM and/or IgG antibody index. cTwenty-two CSF isolates were identified as Borrelia garinii and one as Borrelia afzelii. dNine CSF isolates were identified as B. garinii and one as B. afzelii. eAll three CSF isolates were identified as B. garinii. fThe blood isolate was identified as B. afzelii. gBoth blood isolates were identified as B. garinii. hBoth blood isolates were identified as B. afzelii. iTwenty-six skin isolates were identified as B. garinii and two as B. afzelii. jTwo skin isolates were identified as B. garinii and one as B. afzelii. kFive skin isolates were identified as B. afzelii and one as B. garinii. Borrelial Serologic and Culture Results in Patients With Three Clinical Manifestations of Early European Lyme Neuroborreliosis BS, Bannwarth syndrome; CN, cranial neuritis; CSF, cerebrospinal fluid; EM, erythema migrans; EM-M, erythema migrans-associated meningitis. Data are presented as number (%; 95% CI) or median (IQR). Antibody concentrations were available only in the subgroup tested by chemiluminescence immunoassay (LIAISON). aTwo patients with positive IgM and/or IgG CSF antibodies were seronegative. bAmong 36 patients with positive CSF culture, 28 (78%) had positive serum IgM and/or IgG borrelial antibodies; 26 (72%) had positive CSF IgM and/or IgG borrelial antibodies; 21 (58%) had positive IgM and/or IgG antibody index. cTwenty-two CSF isolates were identified as Borrelia garinii and one as Borrelia afzelii. dNine CSF isolates were identified as B. garinii and one as B. afzelii. eAll three CSF isolates were identified as B. garinii. fThe blood isolate was identified as B. afzelii. gBoth blood isolates were identified as B. garinii. hBoth blood isolates were identified as B. afzelii. iTwenty-six skin isolates were identified as B. garinii and two as B. afzelii. jTwo skin isolates were identified as B. garinii and one as B. afzelii. kFive skin isolates were identified as B. afzelii and one as B. garinii. CSF Findings According to Duration of Neurologic Signs/Symptoms The duration of neurologic signs/symptoms ranged from 1 day to 6 months, with a median of 21 days (IQR 10–35), and was longest in patients with Bannwarth syndrome (Table 1; Figure 2). CSF parameters evolved over time: leukocyte count, protein, and albumin increased during the first 4 weeks and then plateaued, whereas IgM and IgG peaked approximately at week 6 (Figure 3). Duration of neurological symptoms prior to lumbar puncture by clinical manifestation. Distribution of symptom duration (weeks) in patients with Bannwarth syndrome (BS), cranial neuritis (CN), and EM-associated meningitis (M). Temporal evolution of cerebrospinal fluid parameters in early Lyme neuroborreliosis. Changes in CSF leukocyte count, protein, albumin, immunoglobulin concentrations, and corresponding quotients in relation to duration of neurological symptoms. Lines represent trends over time. With longer duration of neurologic involvement, there was a significant increase in the proportion of lymphocytes, the proportion of patients with elevated protein levels, and concentrations of IgM and IgG, as well as IgM and IgG quotients. These associations were largely driven by patients with meningoradiculoneuritis (Supplementary Tables 3A–3D). Multivariate Analysis In the multivariate analysis, age, Bannwarth syndrome, duration of neurologic symptoms, and sex emerged as the main factors significantly associated with CSF characteristics, whereas EM-associated meningitis was not significantly associated with most parameters. Older age was significantly associated with lower CSF leukocyte and lymphocyte counts, as well as with higher glucose levels and a greater likelihood of elevated protein concentrations. Bannwarth syndrome was consistently associated with higher leukocyte and lymphocyte counts, as well as increased levels of total protein, albumin, and IgG, indicating a more pronounced inflammatory response in the CSF. Longer duration of neurologic symptoms was significantly associated with a higher lymphocyte-to-leukocyte ratio and increased concentrations of IgM and IgG (including their CSF/serum quotients), suggesting a time-dependent evolution of the intrathecal immune response. Female sex was associated with lower levels of protein, albumin, immunoglobulins (IgM and IgG), and their quotients, as well as with a lower likelihood of elevated protein, indicating less pronounced CSF abnormalities compared with males. Neutrophil counts and the CSF/serum glucose ratio were not significantly associated with the examined factors (Supplementary Table 4). DISCUSSION In this large single-centre cohort of 405 patients with early European LNB, we identified substantial differences in routine CSF parameters across the three principal clinical manifestations. Meningoradiculoneuritis (Bannwarth syndrome) was associated with the most pronounced inflammatory response and blood–brain barrier dysfunction, whereas EM-associated meningitis exhibited the mildest abnormalities. These findings indicate that early LNB phenotypes differ not only clinically but also quantitatively in intrathecal inflammation. Patients with Bannwarth syndrome consistently showed higher leukocyte counts, elevated protein and albumin concentrations, and increased immunoglobulin levels and quotients, suggesting more extensive inflammatory activity and barrier disruption. In contrast, EM-associated meningitis appears to represent a milder inflammatory phenotype. Although the observed differences are not sufficiently distinct to reliably differentiate clinical phenotypes in individual patients, they provide insight into the biological heterogeneity of early European LNB. The more pronounced inflammatory response observed in patients with meningoradiculoneuritis may reflect more extensive involvement of the nerve roots and meninges, whereas the comparatively mild CSF abnormalities in EM-associated meningitis suggest a less intense intrathecal inflammatory process. The gradient in borrelial serological responses further supports this interpretation. Higher rates of serum and CSF IgG positivity and intrathecal antibody synthesis in Bannwarth syndrome are consistent with a more advanced or immunologically mature stage of infection. Conversely, lower rates of intrathecal antibody production in EM-associated meningitis may reflect either earlier disease or a less intense immune response. We also demonstrate that CSF parameters evolve over time. Cellular and protein changes increase during the first weeks of disease and subsequently plateau, whereas immunoglobulin responses peak later. The association between longer symptom duration and increasing immunoglobulin levels likely reflects progressive intrathecal immune activation and blood–brain barrier impairment. Notably, these temporal patterns were driven primarily by patients with meningoradiculoneuritis. Older age was associated with lower CSF leukocyte counts, possibly reflecting age-related differences in immune responsiveness, although the clinical relevance of this finding remains uncertain. Higher CSF glucose concentrations in older patients may also partly reflect the higher prevalence of impaired glucose metabolism with increasing age. However, because information on diabetic status was not consistently available in this retrospective cohort, this explanation remains speculative. Patients who had received short-term antibiotic treatment before lumbar puncture did not differ significantly from untreated patients with respect to any of the routinely assessed CSF parameters. In addition, the duration of neurologic symptoms before lumbar puncture was virtually identical in the two groups. However, because this was not a longitudinal study, these findings should not be interpreted as evidence that antimicrobial therapy has no effect on the evolution of CSF inflammation. The low sensitivity of Borrelia culture from CSF and blood is consistent with previous (European) studies and underscores the limited role of direct detection methods in routine diagnostics. Higher positivity rates in skin cultures likely reflect greater bacterial burden in cutaneous tissue during early dissemination. This study has several strengths, including a large sample size, consistent diagnostic criteria over two decades, and comprehensive analysis of routine CSF parameters. Nevertheless, several limitations should be acknowledged. The retrospective design may have introduced unrecognized sources of bias, and serological methods changed during the study period. We did not systematically investigate co-infections with other tick-borne pathogens, such as Anaplasma phagocytophilum, although such infections are uncommon in patients presenting with CSF pleocytosis in our setting. In addition, we did not assess CSF cytokines or other markers of immune activation. Finally, although several CSF parameters differed significantly between clinical phenotypes, the substantial overlap of individual values limits their usefulness for distinguishing these manifestations in routine clinical practice. In conclusion, early European LNB is not a uniform entity but a spectrum of clinical phenotypes that differ in the intensity and temporal dynamics of intrathecal inflammation. Meningoradiculoneuritis is associated with the most pronounced intrathecal inflammatory response, followed by cranial neuritis without radicular pain, whereas EM-associated meningitis represents a comparatively milder phenotype. Recognition of these differences may facilitate interpretation of CSF findings and improve understanding of the biological spectrum of early European LNB. The clear association between symptom duration and CSF abnormalities further emphasizes the importance of early diagnostic evaluation, as inflammatory markers evolve over time. Together, these findings refine the clinical framework for diagnosing and staging early LNB. Supplementary Material Notes Financial support. This research was supported by the Slovenian Research and Innovation Agency [grant numbers P3-0296, J3-1744 and J3-8195 to F.S., and P3-0154 to A.K.] (www.aris-rs.si). The funding source had no role in study design, data collection and analysis, interpretation of data, decision to publish, or preparation of the manuscript. Data availability. The data underlying this article will be shared on reasonable request to the corresponding author. Contributor Information Katarina Ogrinc, Department of Infectious Diseases, University Medical Centre Ljubljana, Ljubljana, Slovenia. Petra Bogovič, Department of Infectious Diseases, University Medical Centre Ljubljana, Ljubljana, Slovenia. Andrej Kastrin, Institute for Biostatistics and Medical Informatics, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia. Eva Ružić-Sabljić, Institute of Microbiology and Immunology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia. Franc Strle, Department of Infectious Diseases, University Medical Centre Ljubljana, Ljubljana, Slovenia. Supplementary Data Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author. References Associated Data Supplementary Materials