Case Report
Austin J Clin Neurol 2021; 8(2): 1152.
Evaluation of an Algorithm for the Diagnosis and Therapy of Lyme Neuroborreliosis: A Follow-up Study
Linauer L and Brunner J*
Department of Paediatrics, Medical University Innsbruck, Austria
*Corresponding author: Jurgen Brunner, Department of Paediatrics, Medical University Innsbruck, Anichstrasse 35, A-6020 Innsbruck, Austria
Received: August 04, 2021; Accepted: August 20, 2021; Published: August 27, 2021
Abstract
Background: Although Lyme Neuroborreliosis (LNB) is often seen in paediatric practice, diagnostic criteria for LNB in children are not clearly defined. The guidelines for LNB in adults are based on a combination of clinical picture, CSF pleocytosis and the detection of specific antibodies against Borrelia burgdorferi in CSF and serum. Diagnostic procedure takes several days, thus it isn´t useful in deciding for the need of prompt antibiotic treatment. Aim of study was a retrospective evaluation of an algorithm for the diagnosis and therapy of lyme’s disease, which is used since 2005 at the paediatric department of Innsbruck.
Patients and Methods: All patients presenting with acute peripheral facial palsy from January 2006 to December 2014 were reviewed. The patients were diagnosed according to the algorithm, based on the criteria of the German Society of Neurology. The focus lay on evaluation of diagnosis and therapy according to the algorithm and whether overtreatment and underdiagnosis could therefore be avoided.
Results: 120 patients were enrolled with peripheral facial palsy. 65 (54%) were handled as bell´s palsy and 55 (46 %) as B. burgdorferi s.l. infection. 19 cases were classified as confirmed LNB, 10 as probable and 26 as possible LNB. A total of 69 patients (58 %) were treated correctly according to the algorithm, 16 (13%) were over treated and 14 (11%) under treated with antibiotics. 21 (18%) could not be classified, according to the algorithm, due to the lack of CSF results. Although receiving proper treatment, 3 cases had a persistent defect after recovery.
Conclusions: The algorithm is an appropriate diagnostic tool for the diagnosis and therapy of LNB, particularly with regard to the necessity of a prompt antibiotic treatment, and therefore helpful to avoid underdiagnosis and overtreatment.
Introduction
Lyme borreliosis is the most common cause of tick-borne infection and Lyme Neuroborreliosis (LNB) is the second most common manifestation of infection with Borrelia burgdorferi sensu lato (B. burgdorferi s.l.) in Europe [1,2]. It`s an inflammatory multisystemic disease, which is endemic in the temperate zones of the northern hemisphere [3]. In Tyrol the incidence was calculated 5.2/100000 children/year [2]. The disease evolves in different presentations (Tabelle einfügen) [4]. Although Lyme Neuroborreliosis (LNB) is often seen in paediatric practice, diagnostic criteria for LNB in children have not been established yet. The diagnostic criteria for LNB in adults are based on a combination of clinical picture, Cerebrospinal Fluid (CSF), pleocytosis and the detection of specific antibodies against Borrelia burgdorferi in CSF and serum. Because the diagnostic procedures take several days, they are not helpful in deciding about the necessity of a prompt antibiotic treatment.
Aim of the study was a retrospective evaluation of an algorithm for the diagnosis and therapy of lyme´s disease, which is used since 2005 at the Department of Paediatrics, medical University Innsbruck. This algorithm is based on the diagnostic criteria of LNB by the German Society of Neurology [3].
Patients and Methods
Paediatric patients, diagnosed with acute peripheral facial palsy in the period between 1st of January 2006 and December 31st 2014, were included. As a unique tertiary hospital in tyrol, all paediatric patients with neurologic symptoms are expected to be referred to the Department of Paediatrics in Innsbruck. In addition to the research at Medical University of Innsbruck, our patients were matched to those whose sera and CSF were sent to the Department of Hygiene, Medical Microbiology and Social Medicine. 120 cases of acute peripheral facial palsy were treated at the Department of Pediatrics of the Medical University Innsbruck between January 2006 and December 2014. The mean patient age was 11.16 years (SD 4.17, range 0.10-18) with a slight predominance of male patients (64 to 56 female patients). All cases were evaluated for demographic data, presenting symptoms, clinical examination, CSF-, serological analysis, use of algorithm, antibiotic treatment and clinical outcome. Pleocytosis in the CSF analysis was defined as the appearance of ≥5/μl mononuclear cells. Specific antibodies to B. burgdorferi s.l. (IgG, IgM) in CSF and blood were detected by ELISA, in case of positive or borderline results, samples were confirmed by Western immunoblotting. Restitutio ad integrum means a complete cure, patients with reparatio suffer from persistent defect after healing. Antibiotic treatment is defined as an intravenous single dose of ceftriaxone (Rocephin®) 50mg per kg per day, for 14 days. In cases of overtreatment patients received antibiotics, although watch and wait according to the algorithm was recommended, or the antibiotic treatment lasted longer than 14 days. Undertreatment was detected, if another antibiotic agent or drug than recommended in the guideline was prescribed, or the right one was prescribed for a too short term or at the wrong dosage. The study was authorized by the Ethics Committee of Medical University Innsbruck.
Results
120 patients with acute peripheral facial palsy were diagnosed and treated at the Department of Pediatrics of the Medical University Innsbruck. A tick bite was noticed, by the patient or their parents, in 26 cases and erythema migrans was seen in 3 cases. Peripheral facial palsy due to an infection with Borrelia burgdorferi was seen in 55 patients (46%). Most of them occurred during the summer: 54% from May to August, 30% from September to December and 16% between January and April. In 112 of 120 cases lumbar puncture was performed (93%), the CSF of 42 children showed a pleocytosis with cell counts between 5 to 355 mononuclear cells/μl. Protein content was elevated in 22 patients with a maximum of 5230 mg/L, whereas glucose content was normal in all samples. In 19 cases IgG antibodies against B. burdorferi s.l. were detected by ELISA in the CSF, immunoblotting confirmed this result in 9 patients. 5 samples were tested positive for IgM antibodies and confirmed by immunoblotting in 1 case. The antibody specific index was examined in 11 patients and the median was 4,48 (range 0-35). A PCR for B. burdorferi s.l. was performed 23 patients, ( 2 positive, and 17 negative). In four patients too less CSF was sent to the laboratory. Blood samples were taken from all 120 children with facial palsy. IgG antibodies against B. burdorferi s.l. were tested positive in 30 samples by elisa and 31 samples by immunoblot. IgM antibodies were detected in 31 blood samples and confirmed in 24 via immunoblot. 8 of 114 children suffered from headache and vomiting after lumbar puncture and received symptomatic treatment with paracetamol, ibuprofen or naproxen, considering the age. A total of 69 patients (58 %) were treated correctly according to the algorithm, 16 (13%) were over treated and 14 (11%) under treated with antibiotics. 21 couldn´t be assigned to the algorithm, for lack of lumbar puncture or CSF results, because too less material was sent.
Application of the diagnostic criteria of the DGN classified 19 patients as confirmed, 10 as probable and 26 as possible neuroborreliosis. Of the 19 confirmed, 15 were treated correctly according to the algorithm, 3 patients were undertreated and 1 patient was over treated with ceftriaxone i.v. for 21 days. Half of 10 probable neuroborreliosis cases were treated with i.v. ceftriaxone for 14 days, whereas the other half was undertreated according to algorithm. 9 of 16, who were classified as possible neuroborreliosis were treated correctly, the rest was over-, undertreated or couldn`t be classified according to the algorithm. Despite correct treatment according to the algorithm, this retrospective study showed persistent defects after recovery in 3 of 120 cases (2,5%). 1 male patient at the age of 16 was diagnosed as probable LNB and received 2g ceftriaxone per day for 14 days. Another male patient at the age of 7 was diagnosed as confirmed LNB and was treated with 1500mg ceftriaxone per day for 14 days. And the last one, a male patient at the age of 8 months, was diagnosed as bell´s palsy and received only symptomatic therapy.
Discussion
This is the first study which evaluates a diagnostic and therapeutic algorithm for practical use in LNB. This study is a consecutive follow up study regarding an investigation dealed with LNB in Tyrol, where an algorithm for paediatric LNB was discussed [2]. Consecutively, this algorithm was established at the Department of Paediatrics in Innsbruck 2005.
Stadium 1 (local infection) had rare findings, only 26 of 120 patients or their parents recognised a tick bite and only 3 (2,5 %) an erythema migrans. In contrast, other studies claim erythema migrans to occur at a frequency of >80% as the first manifestation of borrelia infection [1]. So either early manifestations are less common in paediatric patients, or frequency specifications are overestimated. More than half of all cases (53%) presented between may and august, therefore an increased attention and body inspections on children during these months seem reasonable. However, cases of borrelia infection presented throughout the year, so paediatricians should consider infection with Borrelia burgdorferi in any season if patients presented with appropriate symptoms. In 93% of all cases, lumbar puncture was performed, it seems to be one of the most important steps in confirming the diagnosis. In 5 of 8 patients who did not undergo CSF investigation, antibiotic therapy was prescribed on suspicion. Retrospective there was no need for it in all 5 cases. Although over therapy remained without consequences in those 5 cases, the decision-making aid of CSF findings for the need of prompt antibiotic treatment seem to be essential for a structured treatment plan. The ASI is considered to be the most sensitive method for detecting intrathecal antibodies [2]. It was determined in 11 samples and was >2.0 in most cases of confirmed neuroborreliosis. In many other cases ASI could not be tested by the laboratory, because not enough CSF was sent in and ELISA as well as Immunoblot were prioritized. It should be considered to determine ASI in each case of CSF analysis, in order to be able to classify patients according to the algorithm in case of borderline serology results. PCR from liquor samples was only carried out in 23 samples, with 1 positive, 1 weak positive and 17 negative findings. In 4 cases too less liquor was submitted. In addition, in the preliminary study Borrelia DNA could not be detected in any case [2]. The prevalence of positive DNA is reported to be 20%, although it is only recommended to be tested at an early stage, if no antibodies have been produced. Hence, PCR examination should only be considered, if antibodies cannot be detected in early stage of lyme’s disease, but there still remains clinical suspicion [3].
Although only 62 out of 120 patients with facial palsy received an eye patch, this is always recommended for incomplete eyelid closure in terms of prophylaxis for keratoconjuncitivits.
A total of 3 patients were undertreated because i.v. ceftriaxone was switched to oral antibiotics prematurely, only oral antibiotics were given or ceftriaxone was given for too short a time. In all those cases, this didn’t have any consequences in the follow up, but that´s no reason not to stick to the recommendations: of 50 mg/kg/day ceftriaxone intravenously over 14 days [3]. In the medical reports, the reason to change from i.v. to oral regime often was an urge of the children’s parents. It seems to be important to point out the importance of a 14-day intravenous antibiotic therapy.
Based on the algorithm, 19 patients could be classified as confirmed, 10 as probable, 26 as possible cases of neuroborreliosis and 65 patients were hospitalized for bell´s palsy. In comparison to the preliminary study 7 confirmed, 16 probable and 7 possible LNB cases were detected. The remaining facial palsy cases were due to an idiopathic cause. At that time, 4 (6%) of the total 30 seropositive cases were undertreated, without use of a diagnostic algorithm. While 5 (8%) of the 36 cases with Bell`s palsy were antibiotically over treated [2]. In the present study, overtreatment was shown in 16 cases (13%), under therapy in 14 patients (11%) and therapy according to the algorithm in 69 cases (58%). 21 (18%) could not be classified according to the algorithm. In retrospective 13 patients received overtreatment with antibiotics up to 21 days. Contrary our expectations, the number of over- or undertreated patients has not changed significantly with use of the algorithm. However, since the case numbers were relatively low in both studies, especially in the preliminary study, this could also be statistical error. Anyway, many cases were not handled according to the algorithm and probably because of that, some patients were overor undertreated retrospective.
Patient
Age
Sex
Diagnosis
CSF cells
CSF protein
Intrathecal antibodies
PB ELISA
PB Blot
ASI
Antibiotic therapy
Days
Algorithm restitutio ad integrum
ad
1
9
f
Bp
1
155
Neg
Neg
Neg
0
None
pos
2
12
f
pro LNB
5
233
Neg
Neg
IgG, IgM
Cef until results
neg, UT(Grosheva, #19) (Grosheva, #19)
3
9
f
con LNB
0
145
IgG
IgG, IgM
neg
4,9
cef
21
pos
4
15
f
Bp
3
231
neg
IgM
neg
none
pos
5
14
m
Bp
neg
neg
cef
14
nc
6
12
m
Bp
4
192
neg
neg
neg
cef
2
neg OT
7
13
f
Bp
1
146
neg
neg
neg
none
pos
8
15
m
pos LNB
9
238
neg
neg
neg
0
none
neg, UT (Grosheva, #19)
9
11
f
Bp
0
188
neg
neg
neg
none
pos
10
11
f
Bp
2
188
neg
neg
neg
none
pos
11
11
f
pos LNB
2
841
neg
neg
IgG
none
pos
12
8
f
pro LNB
9
212
neg
IgM
IgG, IgM
none
neg, UT(Grosheva, #19) (Grosheva, #19)
13
4
f
con LNB
149
438
IgG
IgG
IgG
cef
14
pos
14
11
m
Bp
1
150
neg
neg
neg
none
pos
15
9
f
Bp
0
270
neg
neg
none
pos
16
17
m
Bp
2
506
neg
neg
neg
none
pos
17
17
f
Bp
1
303
neg
neg
neg
none
pos
18
9
f
Bp
386
neg
neg
neg
cef
14
nc
19
12
f
Bp
703
neg
neg
neg
cef until results
nc
20
15
f
pos LNB
58
355
neg
neg
neg
none
neg, UT
21
15
m
Bp
3
290
neg
neg
neg
none
pos
22
14
m
pos LNB
17
269
neg
neg
none
nc
23
5
f
pro LNB
7
160
neg
IgG, IgM
IgM
cef
21
pos
24
11
m
Bp
3
260
neg
neg
neg
none
pos
25
4
f
Bp
1
129
neg
neg
neg
none
pos
26
10
f
con LNB
218
443
IgG
IgG, IgM
IgG
3,1
cef
14
pos
27
14
m
pro LNB
5
274
neg
neg
neg
cef until results
9
pos
28
16
m
Bp
0
323
neg
neg
neg
none
pos
29
15
m
Bp
3
159
neg
neg
neg
cef until results
neg, OT
30
1
m
con LNB
351
360
IgG
IgG, IgM
neg
cef
14
pos
31
15
f
pos LNB
6
254
neg
neg
neg
cef until results
pos
32
11
f
Bp
0
203
neg
neg
neg
clavamox due to peritonsilarabscess
pos
Table 1: Investigated patients.
33
16
m
pos LNB
5
444
neg
IgM
neg
cefuroxim
28
neg, OT
34
18
f
Bp
neg
neg
none
nc
35
13
f
Bp
217
neg
neg
neg
cef until results
nc
36
14
m
Bp
3
174
neg
neg
neg
none
pos
37
16
f
Bp
1
142
neg
neg
neg
none
pos
38
11
m
Bp
1
247
neg
neg
neg
cef until results
neg, OT
39
8
m
Bp
3
148
neg
neg
neg
none
pos
40
14
f
Bp
1
750
neg
neg
neg
cef until results
neg, OT
41
15
f
Bp
0
171
neg
neg
neg
none
pos
42
14
m
pos LNB
14
391
neg
neg
neg
cef
3
pos
43
8
m
pro LNB
290
439
neg
IgG, IgM
IgG, IgM
cef
14
pos
44
15
f
pos LNB
10
363
neg
neg
cef
7
neg, OT
45
4
m
con LNB
175
508
IgG, IgM
IgG, IgM
IgG, IgM
cef
14
pos
46
13
f
Bp
0
199
neg
neg
neg
none
pos
47
14
m
Bp
neg
neg
neg
none
nc
48
10
f
Bp
1
127
neg
neg
neg
none
pos
49
15
f
Bp
1
272
neg
neg
neg
cef until results
neg, OT
50
10
f
Bp
2
223
neg
neg
neg
none
pos
51
11
m
Bp
4
240
neg
neg
neg
none
pos
52
9
m
Bp
neg
cef and ospamox
nc
53
11
f
pos LNB
neg
IgG
cef
14
nc
54
0,8
m
Bp
neg
neg
none
nc
reparatio
55
15
m
Bp
0
874
neg
neg
neg
none
pos
56
17
f
Bp
1
neg
neg
neg
none
pos
57
4
f
pos LNB
6
neg
neg
neg
none
neg, UT
58
9
f
pos LNB
0
137
neg
IgM
IgM
none
pos
59
3
m
pos LNB
21
335
neg
neg
neg
cef until results
pos
60
13
m
pos LNB
1800
IgG, IgM
IgG, IgM
cef
21
nc
61
11
m
pro LNB
5
250
neg
IgG, IgM
IgM
0
doxycycline
neg, UT
62
14
f
Bp
4
268
neg
neg
neg
none
pos
63
11
m
Bp
3
157
neg
neg
neg
none
pos
64
2
m
con LNB
238
756
IgG, IgM
IgG, IgM
IgG, IgM
cef
14
pos
65
11
m
Bp
neg
neg
none
nc
66
9
m
pro LNB
75
278
neg
neg
IgM
cef
14
pos
67
16
f
Bp
1
372
neg
neg
neg
none
pos
68
17
m
Bp
2
222
neg
neg
neg
none
pos
69
14
m
pos LNB
3
259
neg
IgG
IgG
none
pos
70
15
f
Bp
0
211
neg
neg
neg
none
pos
71
15
m
Bp
neg
neg
none
nc
72
6
f
con LNB
57
1284
IgG
IgG, IgM
IgG, IgM
claforan, cef
3, 14
neg, UT
73
14
m
pos LNB
IgG, IgM
IgG, IgM
cef
14
nc
74
15
m
pos LNB
930
neg
IgM
IgG, IgM
clavamox
nc
Table 1 of 1:
75
16
f
Bp
3
197
neg
neg
neg
none
pos
76
2
m
Bp
1
195
neg
neg
neg
none
pos
77
6
f
pos LNB
2
167
neg
IgM
IgG
amoxicillin
14
neg, OT
78
12
f
Bp
1
217
neg
neg
neg
none
pos
79
11
f
Bp
2
neg
neg
neg
none
pos
80
7
f
pos LNB
18
390
neg
neg
neg
cef until results
pos
81
15
m
pos LNB
8
402
neg
neg
neg
none
neg, UT
82
9
m
pos LNB
IgG, IgM
IgG, IgM
cef
14
nc
83
16
m
pos LNB
6
408
neg
neg
neg
none
neg, UT
84
14
m
con LNB
16
195
IgG
IgG, IgM
IgG
13,7
cef
14
pos
85
8
m
Bp
523000
neg
neg
neg
cef until results
nc
86
16
m
pos LNB
101
378
neg
neg
neg
cef until results
pos
87
6
f
con LNB
110
461
IgG, IgM
IgG, IgM
IgG, IgM
cef
14
pos
88
15
m
Bp
2
476
neg
neg
neg
cef until results
neg, OT
89
7
m
con LNB
355
870
IgG, IgM
IgG, IgM
IgG, IgM
cef
14
pos
reparatio
90
16
m
Bp
1
377
neg
IgG
neg
none
pos
91
10
m
pro LNB
8
270
neg
IgG, IgM
IgG, IgM
0,6
cef after results
14
neg, UT
92
16
m
pro LNB
98
772
neg
IgM
IgG, IgM
cef
14
pos
reparatio
Table 1 of 2:
93
13
m
Bp
0
142
neg
neg
neg
none
pos
94
6
m
con LNB
53
169
IgG
IgG, IgM
IgG, IgM
cef
14
pos
95
10
m
pos LNB
5
172
neg
cef
14
nc
96
10
f
con LNB
64
667
IgG
IgG, IgM
IgG, IgM
cef
14
pos
97
6
f
Bp
0
592
neg
neg
neg
none
pos
98
18
m
Bp
1
466
neg
neg
neg
cef until results
neg, OT
99
8
f
Bp
2
138
neg
neg
neg
none
pos
100
11
m
con LNB
1
214
IgG
IgG, IgM
IgG
none
neg, UT
101
9
m
Bp
3
293
neg
neg
neg
cef until results
neg, OT
102
5
f
pos LNB
35
203
neg
neg
neg
cef
10
neg, OT
103
16
m
Bp
0
1009
neg
neg
neg
cef until results
neg, OT
104
11
m
Bp
264
neg
neg
neg
cef
14
nc
105
7
m
con LNB
264
550
IgG
IgG, IgM
IgG, IgM
35
cef
14
pos
106
16
f
Bp
1
101
neg
neg
neg
none
pos
107
15
m
con LNB
330
757
IgG, IgM
IgG, IgM
IgG, IgM
1,8
cef
14
pos
108
7
f
Bp
0
neg
neg
neg
none
pos
109
14
f
pos LNB
13
271
neg
neg
neg
none
neg, UT
110
7
m
Bp
1
209
neg
neg
neg
cef
14
neg, OT
111
16
f
Bp
2
681
neg
neg
neg
none
pos
112
6
m
pro LNB
133
382
neg
IgG, IgM
IgG, IgM
cef and amoxicillin
21
neg, UT
113
15
m
pos LNB
2
275
neg
IgG
IgG
none
pos
114
6
f
con LNB
113
228
IgG
IgG, IgM
IgG, IgM
2,4
cef
14
pos
115
6
m
Bp
0
neg
neg
neg
none
nc
116
7
f
Bp
3
266
neg
neg
neg
cef until results
neg, OT
117
6
m
Bp
1
153
neg
neg
none
nc
118
15
f
con LNB
118
IgG
IgG, IgM
cef
21
neg, OT
119
5
f
con LNB
72
IgG
IgG, IgM
IgG, IgM
cef
14
pos
120
12
m
con LNB
56
IgG
IgG
IgG
1,5
cef and doxycycline
14
neg, UT
Table 1 of 3:
However, 3 patients who were adequately treated suffered from a defect healing in long-term follow up.
In conclusion, it can be argued that the algorithm for the diagnosis and treatment of Lyme neuroborreliosis represents a suitable diagnostic and therapeutic tool. By taking clear parameters into consideration, the need for a prompt antibiotic treatment can be easily determined. In all cases where the algorithm was taken into account, a comprehensible therapeutic concept was seen which, except in 3 cases, was always curative.
References
- Nau R, Christen HJ, Eiffert H. Lyme disease-current state of knowledge. Dtsch Arztebl Int. 2009; 106: 72-81.
- Brunner J, Moschovakis G, Prelog M, Walder G, Wuerzner R, Zimmerhackl LB. Lyme neuroborreliosis: aetiology and diagnosis of facial palsy in children from Tyrol. Klin Padiatr. 2010; 222: 302-307.
- Deutsche Gesellschaft für Neurologie; Rauer S. Leitlinien für Diagnostik und Therapie in der Neurologie - Neuroborreliose. 2015.
- Stanek G. Durch Zecken übertragbare Krankheitserreger in Mitteleuropa. Wien Klin Wochenschr. 2005; 117: 373-380.
Download PDF