Pediatric Peripheral Facial Nerve Palsy – the Role of Ear Pathology

doi:10.21203/rs.3.rs-4006802/v1

Background: Nearly half of the acquired cases of peripheral facial nerve palsy (PFNP) are of unknown etiology, a condition termed Bell's palsy (BP). PFNP in children can often be associated with acute otitis media (OM). Differences between the clinical course and prognosis in OM-associated PFNP (OM-PFNP) and BP in children are not well characterized.

Methods: Retrospective review of all children hospitalized in a tertiary care children's hospital between January 2006 and December 2020 with a diagnosis of non-traumatic and non-surgery associated PFNP.

Results: Out of 177 children hospitalized with facial nerve palsy, 46 were excluded (central origin: 28; underlying neurologic condition 12; traumatic etiology: 6). A total of 131 cases of PFNP were included in the study, 80 with OM-PFNP and 51 with BP. Children with OM-PFNP were younger at presentation (38.8±50.4 Vs 96.6±74.7 months, p <0.001), presented more frequently during the winter months (32.5% Vs 25.5%, p< 0.05) and had a higher CRP value [median (IQR): 1.3 (0.18-4.2) Vs 0 (0-0) mg/dL, p< 0.005]. There was no significant difference in the corticosteroid use rate (87.5% Vs 88.2%, p=0.9). Four children with OM-PFNP (and none of the BP group) had an additional underlying pathology (petrous bone osteomyelitis, cholesteatoma, and rhabdomyosarcoma).

Conclusions: Children with OM-PFNP are younger at presentation and should be assessed for the presence of a significant associated middle ear/ petrous bone pathology.

Peripheral facial nerve palsy (PFNP) presents with weakness of the facial muscles, loss of facial expression, taste, and corneal sensation (1). The annual incidence of PNFP in children is 5–21/100,000 (2) (3) .

The causes of PFNP are variable and categorized as congenital, either as isolated presentation or as part of multiple congenital malformations, such as in Mobius syndrome, or acquired. Among the acquired causes, nearly half are of unknown etiology, a condition termed Bell's palsy (BP), which is characterized by rapidly progressive facial paralysis, secondary to idiopathic nerve edema (4) (5). It is usually a self-limited condition that presents acutely and progresses over the course of 3–7 days followed by a gradual improvement. The second most common etiology is infectious. Common infectious etiologies include Lyme disease, Herpes simplex and acute suppurative otitis media. Other less common acquired causes include vascular or other compression processes of the facial nerve (6) (7).

Any inflammatory process of the middle ear and mastoid can result in facial nerve paralysis. The facial nerve traverses the fallopian canal to exit the stylomastoid foramen, thus placing the nerve in close proximity to infectious processes of the temporal bone. Inflammation and edema of the nerve in response to infection compromise the vasculature of the nerve because of the rigid fallopian canal, resulting in facial nerve paralysis (8). Even though otitis media is a possible etiology of PFNP, most children suffering from otitis media do not develop PFNP. Data on the differences between otitis media associated PFNP (OM-PFNP) and idiopathic PFNP are limited.

In the present work we aimed to characterize PFNP among hospitalized patients, concentrating on epidemiological and clinical course differences between OM-PFNP and BP.

We retrospectively reviewed all clinical files of children with suspected PFNP admitted to Schneider Children's Medical Center of Israel between January 2006 and December 2020. In our center, hospitalization policy for PFNP was restricted for the non-resolving cases, additional symptoms or return to the emergency room after a previous discharge or non- idiopathic etiology necessitating admission. Inclusion criteria were children aged equal or less than 18 years with a history and physical findings consistent with acute facial nerve paralysis. We excluded patients with underlying neurologic conditions or if the exact nature of facial paresis could not be differentiated as "central" vs peripheral. BP was diagnosed when other etiology was excluded, and treatment response was observed. Acute OM diagnosis was verified by a pediatric otorhinolaryngologist. In case of OM-PFNP, the hospital policy is to perform a paracentesis of the eardrum and obtain a fluid culture on patients without spontaneous drainage.

Data was retrieved from patients' charts including demographics, past medical history, clinical presentation, laboratory findings, imaging studies, ophthalmology and otorhinolaryngology consultations, treatment protocols and outcome.

To evaluate outcome, we collected information from our hospital out-patient clinics database and several community clinics that were accessible by our database. We reviewed children's follow-up visits from neurology, otolaryngology, ophthalmology, and primary care outpatient clinics through December 2020.

Statistics

Patients' characteristics were compared using a chi-square test / Fisher’s exact test for categorical variables and two-tailed T-test/Mann-Whitney U test for continuous variables. Continuous variables were reported as average (and standard deviation); categorical variables were reported as number (and percentage). A p value of < 0.05 was considered significant. The study protocol was approved by the Declaration of Helsinki the local institutional review board (RMC-0779-17).

A total of 177 children were hospitalized during the study period and were reviewed, of whom 28 were diagnosed as central facial nerve palsy and were excluded. We also excluded 12 patients with underlying neurologic conditions such as demyelinating disease, ischemic brain injury and Guillain Barre syndrome and 6 with a traumatic etiology. A total of 133 cases of non-traumatic PFNP were included in the study, 80 with OM-PFNP and 51 with BP. Patient characteristics are detailed in Table 1. Children with OM-PFNP were younger at presentation (38.8 ± 50.4 Vs 96.6 ± 74.7 months, p < 0.001) and had a higher average CRP value [median (IQR): 1.3 (0.18–4.2) Vs 0 (0–0) mg/dL, p < 0.005]. There was no significant difference in the rate of use of systemic corticosteroids (87.5% Vs 88.2%, p = 0.9). Throughout the study period, there has been no change in the annual rate of patient admission due to either BP or OM-PFNP (Fig. 1). However, there were more admitted patients with OM-PFNP, compared to BP patients, in the months December - February (32.5% Vs 25.5% in the remaining months. p < 0.05) (Fig. 2).

Table 1: clinical characteristics on admission. Abx: antibiotics; CRP: C-reactive protein; CS: corticosteroids.

P value	BP (n=51)	OM-PFNP (n=80)
	23 (45.1)	57.5 (46)	Male gender, n (%)
<0.001	96.9±74.7	38.8±50.4	Age (months)
0.015	4.6±2.6	6±3.1	Days of hospitalization
<0.001	15.1	68.7	Max temp >38^oC (%)
<0.005	0.8±2.7	3.8±7.4	CRP (mg/dl)
<0.001	11.3	98.7	Antibiotic treatment (%)
0.87	84.9	87.5	Corticosteroid treatment (%)
<0.001	54.7	22.5	Imaging (%)

In the OM-PFNP group, 60 patients (75%) underwent tympanocentesis, yielding no effusion in 14 of them. Fourteen had purulent otitis media on admission, and out of the rest patients some were transferred from another hospital or had no effusion drained. A positive middle ear fluid culture was documented in 33 patients, including: streptococcus pneumonia (7), diphtheroids (2), Hemophilus (4), pseudomonas spp.(3), staphylococcus aureus (1), group A streptococcus (3), corynbacterium (2), Fusobacterium Necrophorum (1), coagulase negative staphylococcus (1), and in 9 cultures more than 1 pathogen was identified.

Three children also had acute mastoiditis.

Four OM-PFNP cases had an unusual course and were diagnosed eventually with an additional pathology:

1) A 1.5-year-old boy presented with OM-PFNP and mastoiditis. He did not improve with combined antibiotic therapy and developed worsening edema and erythema over the zygoma. CT scan found osteomyelitis of the petrous bone. He underwent mastoidectomy. A complete resolution was observed after 2 weeks.

2) A 7-year-old boy with OM-PFNP, complained on severe unilateral facial pain a month prior to his admission. He completed treatment with corticosteroids, antibiotics, and antiviral acyclovir without identification of a pathogen in middle ear fluid culture. On physical examination a mass was felt on his neck which led to a diagnosis of parapharyngeal rhabdomyosarcoma involving the petrous bone.

3) A 5-year-old boy was admitted for OM-PFNP and treated with antibiotics and systemic corticosteroids. Due to an incomplete resolution, an otolaryngological examination revealed an additional easily bleeding tissue in the right ear canal. Head computerized tomography revealed a tumor in the petrous bone area which was later found to be a rhabdomyosarcoma by biopsy.

4) An 8-year-old girl with OM-PFNP and a persistent suppurating ear was later diagnosed with cholesteatoma. She underwent a combined approach tympano-mastoidectomy.

Among patients with BP, 29 (54%) underwent imaging studies, including ultrasound (US), computed tomography (CT) and magnetic resonance imaging (MRI).

The indications for imaging included recurrent episodes of BP, no improvement or worsening under treatment for a minimum of 5 days, additional neurological symptoms (headache, dizziness, nausea, vision or gait disturbance), presentation under 2 months of age and patients with a family history of premature suspected familial thrombotic tendency. The reasons to perform MRI in 18 patients included: 4 due to recurrence or worsening under treatment, 8 due to an additional neurological finding, such as dystonia, ataxia, papilledema, vision disturbance, abnormal reflexes; 6 due to different reasons (family history of CVA in young age with contraceptive pills treatment, minor head trauma, prematurity with thrombocytopenia, and a patient with malignancy in background, suspected vascular malformation syndrome and young age of presentation: 4 months). Two patients had a brain US due to young age at presentation, with abnormal reflexes on examination and progressing facial asymmetry. Out of 6 patients who underwent a brain CT, 3 had head injury prior to symptoms; 2 with headache and papilledema and one was transferred from another hospital with suspected mastoiditis without otitis. Five patients underwent both MRI and CT − 2 with head injury and headache, 2 with additional neurologic finding and suspicion for central facialis, and one due to young age in presentation.

Long follow up was available for 40 patients, 24 with OM-PFNP and 16 with BP (4.63 ± 5.16 and 4.34 ± 3.76 months respectively) with no significant difference between the groups (P = 0.86). In 13 out of 24 patients with OM-PFNP (54.2%) a complete recovery was seen up to 2 months after presentation. In the rest 11 patients it took 3–12 months until complete resolution. The 16 patients who were classified as BP showed a resolution of symptoms after a period ranging between 2 weeks to 1 year.

We describe a case series of hospitalized children with PFNP. This cohort may represent the more severe PFNP cases, most probably because of under-representation of BP, which is usually not an indication for hospitalization. Such indications may include recurrent cases of facial nerve paresis, young age at presentation or patients with additional neurological manifestation (9). After identifying the different etiologies, we then compared and characterized the 2 major groups of PFNP: BP and those secondary to OM, with OM-PFNP being the most common etiology (60%). This is unlike previous studies (1), where BP was found in 80% of cases. This may be due to a different hospitalization policy between institutions: children with BP can be discharged from the emergency room when there is no associated acute OM or other neurological presentation, especially those who could initiate treatment and follow up through the outpatient clinic. We presume that many children with BP (the most common cause of PFNP in other case series) could have presented to the emergency room or to the primary care physician but did not need hospital admission, as shown in previous studies (13).

In our cohort, children with OM-PFNP showed a complete resolution with treatment, which included antibiotics and steroidal treatment, both in idiopathic and OM-PFNP. Patients with OM-PNF have better outcome under both steroids and antibiotics treatment. (11) Although the disease course in most cases of PFNP is typically self-limited, it may cause significant temporary oral incompetence and an inability to close the eyelid leading to potential eye injury (8) (9). Early initiation of treatment with systemic corticosteroids may minimize progressive edema and nerve damage. Recovery may still occur up to several months after the acute injury and in severe cases recovery may be incomplete (10) (11). It was also found that corticosteroids significantly reduce the frequency of incomplete recovery from BP, and a significant reduction in motor synkinesis during follow-up (12). We found that children with OM-PFNP were younger at presentation than BP, which incidence has been shown to increase with age (14). In our study, most cases were found to be sporadic, with an equal distribution in all age groups (ranging between 1 month to 17 years. Still, OM-PFNP presented more commonly in the Israeli winter months (January-February) in comparison to BP, as expected for a respiratory tract related disease.

Interestingly we found that in a small number of patients who presented with OM- PFNP (and in none of the BP cases, not even after several months of follow up), a significant hidden middle ear or petrous bone pathology was found, which needed careful examination and additional treatment. A similar observation from adults has been reported by Marzo et al (16). This should be kept in mind during patient evaluation especially if there is no clinical improvement under initial appropriate therapy.

The study limitations were related mainly to its retrospective nature. Patient files were collected for those hospitalized, and as such, these are probably the more clinical severe cases. Children discharged from the emergency department were not included in this work. To evaluate long term prognosis and time length for recovery we were limited to relatively small number of patients who were available on our digital database. We assume that the rest of patients had completed their follow up and fully resolved via the community clinics.

Future studies are needed to compare between hospitalized and outpatient treated patients, and to collect more cases of additional hidden pathology. This might help creating a common guideline for when further investigation is needed in PFNP.

Hospitalized children suffering from non-traumatic PFNP usually have BP or OM-PFNP. A cooperation between pediatricians, otolaryngologists and ophthalmologists in disease's evaluation is recommended. Among cases of OM-PFNP that do not respond to initial treatment we should consider a possible underlying petrous bone pathology, given the facial nerve's tortuous pathway. If resolution does not happen in due time, additional radiographic imaging for evaluation may be needed.

Author Contribution

T.C. data collection and wrote manuscript textS.C. and Y.V. data collectionV.H, Y.L, D,U, E.R, O.S and D.L reviewed the manuscript

Funding Declaration

The authors did not receive support from any organization for the submitted work.

Childhood peripheral facial palsy. Karalok ZS, Taskin BD, Ozturk Z, Gurkas E, Koc TB, Guven A. 5, 2018, Childs Nerv Syst., Vol. 34.
Facial palsy: etiology, outcome and management in children. . Jenke AC, Stoek LM, Zilbauer M, Wirth S, Borusiak P. 2011, Eur J Paediatr Neurol., pp. 15(3):209-13.
Outcome of peripheral facial palsy in children - a catamnestic study. Drack FD, Weissert M. 2013, Eur J Paediatr Neurol. , pp. 17(2):185-91.
Facial nerve paralysis. CJ., Danner. 2008, Otolaryngol Clin North Am. , pp. 41:619-32.
Facial nerve palsy in the newborn: incidence and outcome. Falco NA, Eriksson E. 1990 , Plast Reconstr Surg. , pp. 85:1-4.
Pediatric facial nerve paralysis: patients, management and outcomes. Evans AK, Licameli G, Brietzke S, Whittemore K, Kenna M. 2005 , Int J Pediatr Otorhinolaryngol. , pp. 69:1521-8. .
Infective causes of facial nerve paralysis. . Makeham TP, Croxson GR, Coulson S. 2007, Otol Neurotol., pp. 28(1):100-3.
Facial Nerve Paralysis. . Owusu JA, Stewart CM, Boahene K. 2018, Med Clin North Am., pp. 102:1135-1143.
Pediatric facial reanimation: An algorithmic . Deramo, Paul J. Vol. 47, 2020.
Prognostic factors for facial nerve palsy in a pediatric population: A retrospective study and review. Wolfovitz A, Yehudai N, Luntz M. s.l. : Laryngoscope, 2017.
Facial paralysis associated with acute otitis media. Yonamine FK, TUMA J, Silva RF, Soares MC. Braz K otorhinolaryngologol, 2009 Mar-Apr,75(2):228-30.
Facial nerve paralysis secondary to occult malignant neoplasms. Boahene DO, Olsen KD, Driscoll C, Lewis JE, McDonald TJ. 2004 , Otolaryngol Head Neck Surg., pp. 130(4):459-65.
Facial paralysis in children. Reddy S, Redett R. 2015 , Facial Plast Surg., pp. 31:117-22. .
Bell's palsy--is glucocorticoid treatment enough? . Gilden DH, Tyler KL. 2007 , N Engl J Med. , pp. 357:1653-5.
Corticosteroids for Bell's palsy (idiopathic facial paralysis). Madhok VB, Gagyor I, Daly F, Somasundara D, Sullivan M, Gammie F, Sullivan F. 7, s.l. : Cochrane Database Syst Rev., 2016, Vol. 7.
Facial nerve palsy in childhood. . Pavlou E, Gkampeta A, Arampatzi M. 2012, Brain Dev. , p. 34(5):405.
Facial paralysis caused by malignant skull base neoplasms. Marzo SJ, Leonetti JP, Petruzzelli G. Ear Nose Throat J. , 2002 Dec;81(12):845-9.
Volitional type of facial palsy associated with pontine ischaemia. Töpper R, Kosinski C, Mull M. 1995, J Neurol Neurosurg Psychiatry., pp. 58(6):732-4.
Acute Facial Nerve Palsy in Children:Gold standart menegment. Wohrer, D., et al. s.l. : Children, 2022, Vol. 9.

No competing interests reported.

Pediatric Peripheral Facial Nerve Palsy – the Role of Ear Pathology

Status:

Version 1

Abstract

Figures

Introduction

Methods

Statistics

Results

Discussion

Conclusion

Declarations

Author Contribution

References

Additional Declarations

Status:

Version 1