In Europe, human Lyme borreliosis (LB) is caused by several species of Borrelia burgdorferi sensu lato complex including Borrelia afzelii, Borrelia garinii, Borrelia bavariensis, B. burgdorferi sensu stricto (B. burgdorferi), and very rarely Borrelia spielmanii, Borrelia bissettii, Borrelia lusitaniae, and Borrelia valaisiana. While in North America, LB is caused nearly exclusively by B. burgdorferi.1
Lyme borreliae are transmitted to humans by the bite of an infected Ixodes ricinus tick.1 At the site of a tick bite, a localized skin infection solitary erythema migrans (EM) can develop. Subsequently, dissemination of borreliae through blood stream, and possibly through tissue planes can occur, resulting in other manifestations of LB. When borreliae reach the nervous system, manifestations of Lyme neuroborreliosis (LNB) develop.1,2 In children, the main presentations of LNB are meningitis and peripheral facial palsy (PFP).1,3,4 Other clinical manifestations of LNB in children are rare.
In Europe, LNB is confirmed when all 3 of the following criteria are met: neurologic symptoms suggestive of LNB without other obvious reasons, cerebrospinal fluid (CSF) pleocytosis, and intrathecal B. burgdorferi sensu lato specific antibody production.5,6 Detection of B. burgdorferi sensu lato by culture from CSF is definitive evidence of LNB.6,7 However, cultivation of borreliae from CSF is not routinely used due to low sensitivity, slow growth of borreliae, and the need of a specialized laboratory.5
In adult European patients, LNB is mainly caused by B. garinii, followed by B. afzelii and B. burgdorferi.8–11 Information on the etiology of LNB in children in Europe is limited. In Slovenia, a small Central European country with 2 million inhabitants, LB is endemic.12 In the period 2008–2017, the reported incidence rate was between 207 and 337 cases per 100,000 inhabitants.13 High incidence rate and a research-oriented management of patients with LB enabled us to obtain insight into the etiology of LNB in children.
The aims of the current study were (1) to establish the etiology of LNB in children in Slovenia and (2) to compare the demographic, clinical, and laboratory findings in children infected with B. afzelii with those infected with B. garinii.
MATERIALS AND METHODS
The clinical part of the study was carried out at the Department of Infectious Diseases, University Medical Center Ljubljana, while the microbiologic tests were performed at Institute of Microbiology and Immunology, Faculty of Medicine, University of Ljubljana, Slovenia. The data for the current study were obtained prospectively for all of the patients as a part of different studies using the same clinical protocol in the previous years.14–19 Protocols of previous studies (No. 17/04/96, No. 55/04/01, No. 28/02/04, No. 56/04/04, and No. 47/07/04) and of the current study (No. 166/02/13) were approved by the Medical Ethics Committee at the Ministry of Health of the Republic of Slovenia. Informed consent of parents/legal guardians was obtained before enrolment of the patients into the studies.
Children younger than 15 years with clinical presentation suggestive of LNB (patients with clinical syndrome of serous meningitis, PFP, and patients with febrile illness after a tick bite with symptoms suggestive of nervous system involvement) or confirmed LB (multiple EM) admitted to the Department of Infectious Diseases in the period from January 1, 1996, to December 31, 2012, qualified for the studies.
On admission, demographic features, medical history and clinical data were obtained using a questionnaire; detailed physical and neurologic examinations were performed; and basic hematologic and biochemical investigations, and a lumbar puncture (LP) were performed. In CSF, white blood cell count of ≥5 × 106/L was considered as pleocytosis. CSF flow rate and intrathecal total antibody production were defined by criteria of Reiber and Peter.20 As previously described,14 for each patient, concentrations of albumin and immunoglobulins G (IgG), A (IgA) and M (IgM) in blood and CSF were determined, and albumin (QAlb), IgG (QIgG), IgA (QIgA), and IgM (QIgM) CSF/serum quotients were calculated.
Plasma and CSF samples were inoculated into a modified Kelly-Pettenkofer medium8 and were weekly examined for the presence of spirochetes by dark-field microscopy for up to 9 weeks. For species identification, large restriction fragment patterns obtained after breakdown of bacterial DNAs with MluI enzyme were analyzed by pulsed-field gel electrophoresis (MluI-LRFP).21–23
Those patients who had LP performed, had B. burgdorferi sensu lato isolated from CSF, and had species of B. burgdorferi sensu lato identified by MluI-LRFP were enrolled in the present report.
Patients were divided into groups according to the species of B. burgdorferi sensu lato identified by MluI-LRFP. Categoric data were summarized as frequencies (%) and numerical data as medians (interquartile range). Differences in the categoric data were analyzed using χ2 test or Fisher exact test. Differences in the continuous variables were tested with t test, Welch t test, or Mann–Whitney U test as appropriate. The normality assumption and assumption of homoscedasticity (eg, equal variances) were tested with Shapiro–Wilk test and Bartlett test, respectively. Holm’s method was used to adjust P values to control the Type I error. A Holm’s adjusted P value (Padj) of <0.05 was considered statistically significant.
In the study period of 17 years, there were 153 consecutive children younger than 15 years with clinical presentation suggestive of LNB or confirmed LB (multiple EM) who had B. burgdorferi sensu lato isolated from CSF.
In 40/153 (26.1%) patients, the borreliae did not grow well enough to enable species identification by MluI-LRFP. These patients were excluded from our report. In the remaining 113/153 (73.9%) patients, the species of B. burgdorferi sensu lato was identified by MluI-LRFP: 71/113 (62.8%) isolates were typed as B. garinii and 42/113 (37.2%) as B. afzelii.
Demographic features and clinical characteristics of children with LNB are shown in Table 1. Regarding the gender, age of the patients and the duration of the first sign/symptom before admission, the 2 etiologic groups were comparable.
Most cases of LNB occurred between May and August (78/113, 69.0%) compared with period from September to April (35/113, 31.0%, P = 0.012) (Fig. 1). The difference regarding the monthly distribution of patients between the group of LNB caused by B. garinii and B. afzelii was not statistically significant (P = 0.278).
At the initial examination, 87/113 (76.9%) patients reported associated symptoms in the period from the disease onset up to admission; 71/113 (62.8%) patients reported headache and 39/113 (34.5%) reported fever. There were no statistically significant differences between patients with LNB caused by B. garinii, and those in whom B. afzelii was isolated from CSF (Table 2).
Regarding clinical signs at the initial examination, positive meningeal signs were more often found in children infected with B. garinii, while multiple EM was significantly more common in children infected with B. afzelii (Table 3). One patient infected with B. afzelii had bilateral lymphocytoma.
CSF findings in children with LNB are shown in Table 4. One patient from B. garinii and one patient from B. afzelii group had CSF red blood cells >700 × 106 cells/L and were excluded from these findings. For one patient from B. afzelii group, laboratory data on CSF findings were not available. Two of 110 patients (1,8%) had pleocytosis >1000 × 106 cells/L. Children infected with B. garinii had more often lymphocytic predominance and elevated QAlb compared with children infected with B. afzelii.
This single-center study was designed (1) to identify the species of borreliae that cause LNB in children in Slovenia and (2) to determine the influence of B. burgdorferi sensu lato species isolated from CSF on demographic, clinical, and laboratory findings.
In previous years, at the same center, prospective studies have been conducted in children with signs and symptoms suggestive of LB in which the isolation rate of B. burgdorferi sensu lato from CSF has been established. In some of these studies, B. burgdorferi sensu lato species was determined also. In 41/122 (34%) of children treated in hospital due to serous meningitis, Lyme meningitis was diagnosed and 15/122 (12.3%) have B. burgdorferi sensu lato present in CSF.24 In 29/52 (56%) of Slovenian children with acute PFP, LB was established, and in 7/51 (13,7%), B. burgdorferi sensu lato was isolated from CSF. B. garinii was the only identified species (7/7) in these patients.14 55/214 (25.7%) of children with multiple EM have pleocytosis and 8/199 (4%) of them have B. burgdorferi sensu lato present in CSF. In these patients, B. afzelii and B. garinii were determined in 5/8 and 1/8 patients, respectively, and in 2/8 patients borreliae were undetermined.15 Tick-borne illness was diagnosed in 28/53 (53%) of children with febrile illness after tick bite, and 13/53 (24.5%) of them had LB. Furthermore, 4/53 (7.6%) of these patients had B. burgdorferi sensu lato isolated from CSF.16
In the current study, in the period of 17 years, there were 153 children who had B. burgdorferi sensu lato isolated from CSF and 113/153 had B. burgdorferi sensu lato species identified. These 113 children were included in the present report. In 71/113 (62.8%) and 42/113 (37.2%) children, B. garinii and B. afzelii, respectively, were isolated from CSF. Because they were all consecutive patients treated in one hospital, it can be deduced that in this area LNB in children is most frequently caused by B. garinii, followed by B. afzelii.
When stating that B. garinii causes LNB in our set of patients more often than B. afzelii, questions about culturing of borreliae arise. The cultivation rate of borreliae from CSF is 10%–30%,25,26 and it is not clear whether sensitivity of culture for different subspecies of borreliae is the same.25 Furthermore, once the patient’s specimen reaches the laboratory, contamination is possible.25 However, B. burgdorferi sensu lato culture from patient’s material is still the gold standard for specificity in the laboratory diagnosis of LB.6 The isolation of borreliae from CSF and species identification for the current study was performed at a well-established, specialized, and research-oriented laboratory, where B. burgdorferi sensu lato culture for previous studies in our center was also performed.9,11,27,28 Moreover, in the study period, there was one positive CSF culture in children every 5–6 weeks on average only. Thus, contamination of these specimens is possible of course, but we believe it to be very unlikely.
Our results support the findings of Strle et al11 at the same center of consecutive adult patients with LNB, where B. garinii was isolated in 23/36 (63.9%), B. afzelii in 10/36 (27.8%) and B. burgdorferi in 3/36 (8.3%) CSF specimens. They also support findings of other studies at the same center where B. garinii was more often than B. afzelii isolated from CSF as a cause of LNB,8–10 and 2 European studies where the most frequently detected borreliae in CSF by polymerase chain reaction was B. garinii.29,30 In our series, B. burgdorferi was not isolated from CSF in any of the patients. But, B. burgdorferi was isolated from CSF in the above mentioned study by Strle et al11 and detected in other European studies.22,28,30
Most cases of LNB in children in our study occurred from May to August and less from September to April. This supports the findings of other European studies.14,31–35 Our patients reported tick bites with similar frequency (42.5%) as patients in other European studies (35%–77%).31–36 There was no predilection as to the site of the tick bite. This does not corroborate the findings of the Swedish study by Södermark et al,34 where in 72% of children with LNB, the location of the bite was in the head and neck area, in 24% it was below the neck, and in 4% it was of both regions. On the ground of the location of the tick bite, we could not differentiate patients with LNB caused by B. garinii from patients with LNB caused by B. afzelii.
The patients came to our attention after 4 days of median duration of the first sign/symptom of LNB and that is earlier than was reported in other European studies (median 7–14 days),33,34,37 where positive B. burgdorferi sensu lato CSF culture was not one of the inclusion criteria. This difference can be explained by the fact that we included patients with positive CSF culture only and more successful isolations from patients with early disease (EM or early LNB).25 To a lesser extent, the relatively short duration of the first sign/symptom in our series could be due to strong awareness of the disease or fast availability of pediatric infectious diseases specialist in our country.
Clinical picture of LNB in children was well described in several European studies. Children with LNB in Europe have PFP in 59%–71%, headache in 50%–69%, neck pain/stiffness in 29%–46%, fever in 19%–46%, fatigue in 39%–86%, malaise 26%–52%, loss of appetite 18%–60%, EM in 13%, lymphocytoma in 18%, gastrointestinal symptoms in 23%, loss of weight in 18%, arthralgia in 14%, back pain or pain in extremities in 11%, sensory disturbances in 9%, and behavioral changes in 6%–12%.33,34,36–38 Our group of patients is a rather select one, and we have not performed a comparison of our results with all cases of LNB in children. As such, our results do not describe clinical picture of LNB in children in general. Thus, we cannot compare them with these mentioned studies.
However, our results do offer a comparison of clinical picture of LNB in children caused by B. garinii, to those caused by B. afzelii, which are scarce comparisons in literature. The findings of Strle et al11 show that in cases of early LNB in adult patients, B. garinii causes more typical disease (Bannwarth syndrome), and that LNB caused by B. afzelii is less specific. In our study, patients infected with B. garinii did not report symptoms suggestive of CNS involvement (headache, neck pain, vomiting, nausea, irritability, drowsiness, behavioral changes, dizziness, and visual symptoms) or any other symptoms more often than patients infected with B. afzelii. During the clinical examination in patients infected with B. garinii, except for positive meningeal signs (69.0% vs. 38.1%), other clinical signs suggestive of CNS involvement (PFP, tremor, and voiding disturbances) were not more frequent. Patients infected with B. afzelii had more frequent multiple EM (45.2% vs. 18.3%). In our series, none of the patients had Bannwarth syndrome. Therefore, our results show that the clinical picture of LNB in children caused by B. garinii is not more significant of CNS involvement. These are the main findings of our study and very novel results. The difference with the results of Strle et al11 could be due to the fact that our study included children only, and LNB in children differs from disease presentation in adults.36,39 Our findings do support the very recent findings of Barstad et al29 who found out that children with LNB caused by B. garinii did not have a distinct clinical picture.
In children with LNB, European studies report of median pleocytosis of 129–378 × 106 cells/L34,36,37 and of mononuclear predominance with median values of 97%–98%.31,34 European authors report of elevated CSF albumin (median 293 mg/L)34 and elevated CSF protein values (median 45–54 mg/dL)31,32 in 66.6%32 of children with LNB. Our results support these findings. Additionally, our CSF laboratory results show more frequent lymphocytic predominance (97.1% vs. 75.0%) and more frequent dysfunction of blood–brain barrier (elevated QAlb 80.6% vs. 50.0%) in children with LNB caused by B. garinii compared with LNB caused by B. afzelii. Patients with B. garinii infection had higher median CSF white blood cell count compared with patients with B. afzelii infection (253.5 × 106 cells/L vs. 93.5 × 106 cells/L, Padj 0.323), but the difference was not statistically significant. Similarly, in adult patients, Strle et al11 found out that patients with B. garinii isolated from CSF had significantly higher median CSF white blood cell values, more frequent lymphocytic predominance, and more frequently elevated CSF protein concentrations. Therefore, our results support their findings, only that the difference between the children infected with B. garinii to children infected with B. afzelii is less pronounced. This could be due to the same median duration of symptoms before LP in our 2 groups of patients (4 days) and much shorter duration as in the study by Strle et al.11 In their study, this period for patients infected with B. afzelii was 7.5 months and for patients infected with B. garinii was 19 days, but the difference was not statistically significant. Thus, inflammation seems to be less pronounced in both children and adults with B. afzelii compared with patients with B. garinii isolated from CSF.
The current study is monocentric and thus the epidemiologic findings stand for the central part of Slovenia only. Another limitation is the long study period during which there could be some micro changes in clinical assessment of patients and/or undetectable changes in laboratory environment/procedures that could unknowingly affect our results.
The results of our study present ideas for further research. Due to the low percentage of children with typical features of LNB in the current study (30.1% of patients had PFP and 28.3% had multiple EM), patients with positive CSF B. burgdorferi sensu lato culture who were without PFP and/or multiple EM could be compared with patients with serous meningitis who have very similar presentation. This would be of clinical importance because LNB demands antibiotic treatment. Since the inflammation in CSF is more pronounced in children with B. garinii, but with no really distinct clinical picture of CNS infection, other factors, maybe host factors, influencing clinical presentation should be looked for. Thus, humoral immune response in children with LNB could be described.
In conclusion, this is the largest study in Europe and North America in children with positive CSF B. burgdorferi sensu lato culture. In Slovenia, LNB in children is more often caused by B. garinii followed by B. afzelii. The clinical picture of LNB in children caused by B. garinii is not more significant of CNS involvement compared with children infected with B. afzelii. However, the CSF laboratory results show more pronounced inflammation in children infected with B. garinii compared with children infected with B. afzelii.
1. Stanek G, Wormser GP, Gray J, et al. Lyme borreliosis. Lancet. 2012;379:461–473.
2. Wormser GP. Hematogenous dissemination in early Lyme disease. Wien Klin Wochenschr. 2006;118:634–637.
3. Esposito S, Bosis S, Sabatini C, et al. Borrelia burgdorferi infection and Lyme disease in children
. Int J Infect Dis. 2013;17:e153–e158.
4. Sood SK. Lyme disease in children
. Infect Dis Clin North Am. 2015;29:281–294.
5. Mygland A, Ljøstad U, Fingerle V, et al; European Federation of Neurological Societies. EFNS guidelines on the diagnosis and management of European Lyme neuroborreliosis
. Eur J Neurol. 2010;17:8–16, e1.
6. Stanek G, Fingerle V, Hunfeld KP, et al. Lyme borreliosis: clinical case definitions for diagnosis and management in Europe. Clin Microbiol Infect. 2011;17:69–79.
7. Steere AC, Strle F, Wormser GP, et al. Lyme borreliosis. Nat Rev Dis Primers. 2016;2:16090.
8. Ruzić-Sabljić E, Maraspin V, Lotric-Furlan S, et al. Characterization of Borrelia burgdorferi sensu lato strains isolated from human material in Slovenia. Wien Klin Wochenschr. 2002;114:544–550.
9. Cerar T, Ogrinc K, Cimperman J, et al. Validation of cultivation and PCR methods for diagnosis of Lyme neuroborreliosis
. J Clin Microbiol. 2008;46:3375–3379.
10. Ogrinc K, Lotrič-Furlan S, Maraspin V, et al. Suspected early Lyme neuroborreliosis
in patients with erythema migrans. Clin Infect Dis. 2013;57:501–509.
11. Strle F, Ruzić-Sabljić E, Cimperman J, et al. Comparison of findings for patients with Borrelia garinii and Borrelia afzelii isolated from cerebrospinal fluid. Clin Infect Dis. 2006;43:704–710.
12. Strle F. Lyme borreliosis in Slovenia. Zentralbl Bakteriol. 1999;289:643–652.
14. Arnez M, Ruzić-Sabljić E. Lyme borreliosis and acute peripheral facial palsy in Slovenian children
. Pediatr Infect Dis J. 2010;29:182–184.
15. Arnez M, Pleterski-Rigler D, Luznik-Bufon T, et al. Children
with multiple erythema migrans: are there any pre-treatment symptoms and/or signs suggestive for central nervous system involvement? Wien Klin Wochenschr. 2002;114:524–529.
16. Arnez M, Luznik-Bufon T, Avsic-Zupanc T, et al. Causes of febrile illnesses after a tick bite in Slovenian children
. Pediatr Infect Dis J. 2003;22:1078–1083.
17. Arnez M, Pleterski-Rigler D, Ahcan J, et al. Demographic features, clinical characteristics and laboratory findings in children
with multiple erythema migrans in Slovenia. Wien Klin Wochenschr. 2001;113:98–101.
18. Arnez M, Ruzić-Sabljić E, Ahcan J, et al. Isolation of borrelia burgdorferi sensu lato from blood of children
with solitary erythema migrans. Pediatr Infect Dis J. 2001;20:251–255.
19. Arnez M, Radsel-Medvescek A, Pleterski-Rigler D, et al. Comparison of cefuroxime axetil and phenoxymethyl penicillin for the treatment of children
with solitary erythema migrans. Wien Klin Wochenschr. 1999;111:916–922.
20. Reiber H, Peter JB. Cerebrospinal fluid analysis: disease-related data patterns and evaluation programs. J Neurol Sci. 2001;184:101–122.
21. Belfaiza J, Postic D, Bellenger E, et al. Genomic fingerprinting of Borrelia burgdorferi sensu lato by pulsed-field gel electrophoresis. J Clin Microbiol. 1993;31:2873–2877.
22. Busch U, Hizo-Teufel C, Boehmer R, et al. Three species of Borrelia burgdorferi sensu lato (B. burgdorferi sensu stricto, B afzelii, and B. garinii
) identified from cerebrospinal fluid isolates by pulsed-field gel electrophoresis and PCR. J Clin Microbiol. 1996;34:1072–1078.
23. Ruzić-Sabljić E, Zore A, Strle F. Characterization of Borrelia burgdorferi sensu lato isolates by pulsed-field gel electrophoresis after MluI restriction of genomic DNA. Res Microbiol. 2008;159:441–448.
24. Leskovec A. Pogostnost lymskega meningitisa pri otrocih s sindromom seroznega meningitisa (master’s thesis, COBISS.SI-ID: 24089049). 2008:Ljubljana, Slovenia: Faculty of Medicine, University of Ljubljana; 51.
25. Aguero-Rosenfeld ME, Wang G, Schwartz I, et al. Diagnosis of lyme borreliosis. Clin Microbiol Rev. 2005;18:484–509.
26. Wilske B, Fingerle V, Schulte-Spechtel U. Microbiological and serological diagnosis of Lyme borreliosis. FEMS Immunol Med Microbiol. 2007;49:13–21.
27. Arnež M, Ružić-Sabljić E. Borrelia burgdorferi sensu lato bacteremia in Slovenian children
with solitary and multiple erythema migrans. Pediatr Infect Dis J. 2011;30:988–990.
28. Ruzić-Sabljić E, Lotric-Furlan S, Maraspin V, et al. Analysis of Borrelia burgdorferi sensu lato isolated from cerebrospinal fluid. APMIS. 2001;109:707–713.
29. Barstad B, Quarsten H, Tveitnes D, et al. Direct molecular detection and genotyping of borrelia burgdorferi sensu lato in cerebrospinal fluid of children
with Lyme neuroborreliosis
. J Clin Microbiol. 2018;56:e0186817.
30. Ornstein K, Berglund J, Bergström S, et al. Three major Lyme borrelia genospecies (Borrelia burgdorferi sensu stricto, B. afzelii
and B. garinii
) identified by PCR in cerebrospinal fluid from patients with neuroborreliosis in Sweden. Scand J Infect Dis. 2002;34:341–346.
31. Tuerlinckx D, Bodart E, Jamart J, et al. Prediction of Lyme meningitis based on a logistic regression model using clinical and cerebrospinal fluid analysis: a European study. Pediatr Infect Dis J. 2009;28:394–397.
32. Millner MM, Müllegger RR, Spork KD, et al. Lyme borreliosis of central nervous system (CNS) in children
: a diagnostic challenge. Infection. 1991;19:273–278.
33. Broekhuijsen-van Henten DM, Braun KP, Wolfs TF. Clinical presentation of childhood neuroborreliosis; neurological examination may be normal. Arch Dis Child. 2010;95:910–914.
34. Södermark L, Sigurdsson V, Näs W, et al. Neuroborreliosis in swedish children
: a population-based study on incidence and clinical characteristics. Pediatr Infect Dis J. 2017;36:1052–1056.
35. Peltomaa M, Saxen H, Seppälä I, et al. Paediatric facial paralysis caused by Lyme borreliosis: a prospective and retrospective analysis. Scand J Infect Dis. 1998;30:269–275.
36. Skogman BH, Croner S, Nordwall M, et al. Lyme neuroborreliosis
: a prospective study of clinical features, prognosis, and outcome. Pediatr Infect Dis J. 2008;27:1089–1094.
37. Skogman BH, Glimåker K, Nordwall M, et al. Long-term clinical outcome after Lyme neuroborreliosis
in childhood. Pediatrics. 2012;130:262–269.
38. Christen HJ, Hanefeld F, Eiffert H, et al. Epidemiology and clinical manifestations of Lyme borreliosis in childhood. A prospective multicentre study with special regard to neuroborreliosis. Acta Paediatr Suppl. 1993;386:1–75.
39. Øymar K, Tveitnes D. Clinical characteristics of childhood Lyme neuroborreliosis
in an endemic area of northern Europe. Scand J Infect Dis. 2009;41:88–94.