Facial nerve palsy and Lyme neuroborreliosis in children : treatment and outcome

<p dir="ltr">Background</p><p dir="ltr">Acute peripheral facial nerve palsy is a condition where the muscles on one side of the face become weak or paralysed due to impaired function of the facial nerve. Facial palsy affects both adults and children worldwide, causing difficulties portraying emotions with facial movement, which can significantly impact the social well-being of affected individuals. Studies have shown that 20% of children experience incomplete recovery of facial palsy.</p><p dir="ltr">Idiopathic facial palsy is the most common cause of facial palsy in children in a global perspective. Lyme neuroborreliosis (LNB), a tick-borne infection, is another common cause of paediatric facial palsy, causing up to 65% of cases in children in Borrelia-endemic areas.</p><p dir="ltr">LNB-associated facial palsy is treated with antibiotics with either oral doxycycline or intravenous ceftriaxone, depending on the children's age. Idiopathic facial palsy in adults is treated with corticosteroids which improves and hastens recovery, but there is a lack of evidence of the efficacy of corticosteroids in children with facial palsy.</p><p dir="ltr">Various methods are available to evaluate the physical impairment and social impact of facial palsy. These include physician-assessed facial nerve grading systems and patient/guardian-reported questionnaires. Additionally, neurophysiological examinations may be performed to evaluate the function of the facial nerve.</p><p dir="ltr">Objectives</p><p dir="ltr">The aim of this thesis was to investigate the aetiology, incidence, epidemiology and long-term outcomes of facial palsy in children in the Borrelia-endemic area of Stockholm, Sweden. The effectiveness of oral versus intravenous antibiotic treatment for LNB was compared with regard to clinical recovery. Furthermore, the detectability of brain-damage markers in serum in children with LNB was investigated, as well as their prognostic value in clinical recovery.</p><p dir="ltr">Material and methods</p><p dir="ltr">Studies I and IV are historical descriptive cohort studies with prospectively collected data. The study populations consisted of children diagnosed with acute peripheral facial nerve palsy at the Astrid Lindgren Children's Hospital in the Borrelia-endemic area of Stockholm, Sweden. The study periods were 2014- 2015 for study I and 2016-2020 for study IV. Study participants were evaluated with Sunnybrook Facial Grading System, House-Brackmann Facial Nerve Grading System, Facial Clinimetric Evaluation Scale, Facial Disability Index, Synkinesis Assessment Questionnaire and neurophysiological examinations consisting of electromyography and electroneurography.</p><p dir="ltr">Study II is a historical cohort study that compares the effectiveness of oral doxycycline and intravenous ceftriaxone in children with LNB in regard to clinical recovery. The study population consisted of children with LNB from previously conducted prospective studies. Clinical outcome was compared between the treatment groups receiving either oral doxycycline or intravenous ceftriaxone.</p><p dir="ltr">Study III is a historical case-control study that evaluates the detectability of the brain-damage markers neuron-specific enolase and S100 calcium-binding protein B in children with LNB and non-LNB controls. Children with LNB were identified as cases with age- and gender matched non-LNB children serving as controls. The detectability of the brain-damage markers was evaluated and compared between children with LNB and non-LNB controls. Among children diagnosed with LNB, levels of brain damage markers were compared between those who showed clinical recovery and those who did not.</p><p dir="ltr">Results</p><p dir="ltr">Idiopathic facial palsy and LNB were the two most common causes of facial palsy in children in the studies I and IV. During 2014-2015 LNB caused 58,4% of facial palsies in children and idiopathic facial palsy accounted for 36,4% of cases. During 2016-2020 idiopathic facial palsy and LNB caused 41,7% and 41,2% of cases, respectively. The estimated incidence for these conditions were 12/100.000 children/year for idiopathic facial palsy and 11,8/100.000 children/year for LNB (studies I and IV).</p><p dir="ltr">Facial palsy associated with LNB predominantly affected children under the age of ten, with most cases occurring between June and November in both study periods. After a follow-up period of three months, 8% of the children had incomplete recovery of their facial palsy (study I). When evaluating the long- term outcome of idiopathic facial palsy, differences were found in the evaluation according to the methods applied (grading systems, questionnaires and neurophysiological examination) (study IV).</p><p dir="ltr">Children with LNB showed no difference in clinical outcome when comparing treatment with oral doxycycline and intravenous ceftriaxone when adjusted for relevant variables (study II). The serum levels of the brain damage markers evaluated did not differ significantly when comparing children with LNB and non- LNB controls. The serum levels did not differ when comparing children with complete recovery and incomplete recovery, among those diagnosed with LNB (study III).</p><p dir="ltr">Conclusion</p><p dir="ltr">Idiopathic facial palsy and LNB are the main causes of facial palsy among children in the Borrelia-endemic area of Stockholm, Sweden. LNB mainly affects children under the age of ten during June to November. Children may have persistent long-term symptoms of facial palsy, with a discrepancy seen between the evaluation according to physician-assessed grading systems, patient/guardian-reported questionnaires and neurophysiological examinations. Treatment of LNB in children with oral doxycycline seems to be equivalent to intravenous ceftriaxone in regard to clinical recovery. The brain damage markers NSE and S100B where equally detectable in children with LNB and non-LNB controls and cannot be recommended as prognostic biomarkers for clinical recovery in children with LNB.</p><h3>List of scientific papers</h3><p dir="ltr">I. Peripheral facial nerve palsy in children in a Borrelia high-endemic area, a retrospective follow-up study. <b>Arnason S</b>, Hultcrantz M, Nilsson A, Laestadius Å. Acta Paediatrica. 2020;109(6):1229-35. <a href="https://doi.org/10.1111/apa.15063" rel="noreferrer" target="_blank">https://doi.org/10.1111/apa.15063</a></p><p dir="ltr">II. Effectiveness of antibiotic treatment in children with Lyme neuroborreliosis - a retrospective study. <b>Arnason S</b>, Skogman BH. BMC Pediatr. 2022 Jun 9;22(1):332. <a href="https://doi.org/10.1186/s12887-022-03335-w" rel="noreferrer" target="_blank">https://doi.org/10.1186/s12887-022-03335-w</a></p><p dir="ltr">III. Brain damage markers neuron-specific enolase (NSE) and S100B in serum in children with Lyme neuroborreliosis - detection and evaluation as prognostic biomarkers for clinical outcome. <b>Arnason S</b>, Molewijk K, Henningsson AJ, Tjernberg I, Skogman BH. Eur J Clin Microbiol Infect Dis. 2022 Jul;41(7):1051-1057. <a href="https://doi.org/10.1007/s10096-022-04460-1" rel="noreferrer" target="_blank">https://doi.org/10.1007/s10096-022-04460-1</a></p><p dir="ltr">IV. Aetiology of pediatric facial nerve palsy and long-term outcome of idiopathic facial palsy. <b>Arnason S</b>, Upate Z, Laestadius Å, Hultcrantz M, Skogman BH, Marsk E. Submitted for publication in September 2025. [Manuscript]</p>