Differential diagnosis of peripheral facial nerve palsy: a retrospective clinical, MRI and CSF-based study.

BACKGROUND: Facial nerve palsy is the most common cranial nerve disorder. There is no consensus on a single diagnostic tool deemed as the 'gold standard' for distinguishing between idiopathic (Bell's palsy) and symptomatic causes. The diagnosis is one of exclusion and most often made on physical examination. In the present study, we describe the etiological background of peripheral facial palsy in N = 509 patients and evaluate the relevance of cerebrospinal fluid (CSF) analysis and magnetic resonance imaging (MRI) in differential diagnosis.

METHODS: We carried out a retrospective data analysis of 509 patients with the clinical diagnosis of peripheral facial palsy admitted to our emergency unit between January 2006 and January 2017. All patients were seen clinically; their CSF was analyzed and MRI was performed.

RESULTS: Of N = 526 patients with isolated facial palsy, 17 patients were excluded because they did not consent to CSF analysis. Of the remaining N = 509 patients, 383 patients (75.2%) were diagnosed with idiopathic facial palsy. In the remaining 126 patients (24.8%), the following etiologies for facial palsy could be found: Ramsay-Hunt-Syndrome (N = 34), Lyme Neuroborreliosis (N = 32), other viral/bacterial central nervous system (CNS) infections (N = 8), neoplasias (N = 18), autoimmune disease (N = 12), otogenous processes (N = 6), or other etiologies (N = 16). Analysis of the CSF showed 85% sensitivity for Ramsay-Hunt-Syndrome and 100% for Lyme Neuroborreliosis and other viral/bacterial CNS infections. CSF analysis proved a reliable diagnostic tool for identifying these subgroups. MRI with contrast compounds, as performed in 409 patients, was the most important tool in diagnosing neoplasias (88% sensitivity) and otogenous processes (83% sensitivity). MRI with contrast-enhancing compounds did not reveal additional information concerning inflammatory facial nerve lesions when performed the same day as hospital admission.

CONCLUSIONS: Although peripheral facial palsy was predominantly idiopathic (75.3%) in our cohort, the disease was caused in approximately 25% of the patients by factors which require specific treatment. In the present study, CSF analysis proved to be the leading method for the diagnosis of Ramsay-Hunt-Syndrome, Lyme Neuroborreliosis, and other CNS infections. These subgroups made up approximately 15% of our cohort. To detect these subgroups reliably, routine use of CSF analysis in peripheral facial palsy may be advisable, whereas MRI proved to be useful for exclusion of otogenic and neoplastic processes with a sensitivity of 83% and 88%. We found that the use of MRI with contrast-enhancing compounds does not provide additional diagnostic information on the day of hospital admission. Hence, the potential benefits of routine use of MRI in patients with facial nerve palsy should be weighed against health care cost factors.