OHDSI Home | Forums | Wiki | Github

Phenotype submission - Facial Palsy (LMN) including Bells Palsy thats not UMN

Facial Palsy including Bells Palsy (LMN thats not UMN)

Clinical Description

Authoritative source:

Facial Palsy https://www.ncbi.nlm.nih.gov/books/NBK549815
Bells Palsy https://www.ncbi.nlm.nih.gov/books/NBK482290/

Abstracted from authoritative source:

Overview: Facial nerve palsies are a common, and can affect the intracranial, intratemporal, and extratemporal course of its branches each with different presentation. The motor function of the peripheral facial nerve controls the upper and lower facial muscles.
Bell palsy (BP) is the most common peripheral paralysis of the seventh cranial nerve with an onset that is rapid and unilateral lower motor neuron (LMN) without preservation of forehead muscle function. If forehead strength is preserved, a central cause upper motor neuron (UMN)of weakness should be considered. UMN is considered stroke, and is not part of this phenotype.

Most commonly (70%), the cause for facial nerve palsy remains unknown and has the name ‘Bell palsy.’ Bell palsy, a diagnosis of exclusion, has an incidence of 10 to 40 per 100000. The annual incidence is 15 to 20 per 100,000 with 40,000 new cases each year and the lifetime risk is 1 in 60. There is an 8% to 12% recurrence rate. It is a diagnosis of exclusion, onset within 24-48hrs and may take upto 1 year for recovery. Other causes include Trauma (10-20%) of temporal skull bone (hemotympanum, battle sign, nystagmus), surgery of ear canal; viral (5%) - Herpes Zoster infection resulting in facial paralysis due to geniculate ganglionitis (also known as Ramsay Hunt syndrome (RHS). Neoplasias (2%) - parotid neoplasm, acoustic neuroma,

Presentation: Bells palsy most common, rapid and suddent onset. Other causes may have characteristic symptoms such as trauma, pain. Upper motor neuron (stroke) may have charcteristic stroke symptoms in rest of body (e.g. unilateral hemiplegia)

Diagnostics Evaluation: While the majority of cases are found to be idiopathic, any clinician needs to rule out a cerebrovascular event or other serious underlying pathology before diagnosis Bell’s palsy.

Therapy Plan: Bell’s palsy - the use of steroids and analgesia is though to increase recovery of motor function if started within 72 hours of symptom onset. If, within this period, prednisolone 50 mg once a day should be commenced for ten days. For Ramsay Hunt Syndrome, add acyclovir 800 mg 5 times a day. For certain bells palsy - surgical decompression of the facial canal. Transcutaneous Nerve Stimulation Transcutaneous nerve stimulation is an additional new treatment option for those with unilateral facial nerve palsy. The technology uses EMG signals from muscles on the intact side of the face to simultaneously stimulate the corresponding muscles on the side of paresis.

Prognosis: Bells palsy - Around 1 year for recovery. May relapse in about 10%. Even without treatment, 70% of patients will have complete resolution.

Differential Diagnosis: upper motor neuron stroke (exclusion)

Regular Expression: weakness|tinnitus|ear|virus|viral|stroke

Phenotype Development:

Logic Description: All events of Facial Palsy with no events in the prior 183 days clean window, with no evidence of congenital facial palsy and no recent upper motor neuron disease. A person once diagnosed is assumed to suffer from Facial palsy for a minimum of 6 months. No new events are allowed within 183 days (half year) of prior event.

Source of errors in real world and impact on algorithm:

  • Miss rate/False negative rate - we hypothesize that it is less likely that this condition would be missed when present because of its characteristic presentation and relatively rapid onset.
  • Index date misclassificaiton: we hypothesize that because most bell’s palsy presents itself within 48hours as a full blown facial palsy, it is not an indolent disease - and so the probability of index date misclassificaiton is lower.
  • Specificity - since isolated upper motor neuron stroke of the facial nerve nucleus is less common, and if UMN it is more likely from unilateral hemiplegia - we believe it is unlikely that a person with UMN will be diagnosed as a LMN

Cohort Submission:

This cohort definition has cohort id # 256 in OHDSI Phenotype library (pending peer review).

Insights from Cohort Diagnostics:

  • Impact of rule to remove UMN etiology of facial palsy: in almost all data sources the impact was <1% except in 1 data source where the impact was 4%.
  • Incidence rate: No inconsistent rate differences observed over time and by gender for both definitions however it seems to become more prevalent as age increases. Most data sources had similar incidence rate and patterns.
  • Index event breakdown: Bell’s palsy accounts for nearly 99% of person entry events. With facial palsy being < 1%. This pattern was observed in all but one data source (an ambulatory electronic health record system in USA). Reason for this is unknown.
  • Visit context: in data sources that has good capture of inpatient/er visit - we observed that about 30% of persons had a visit in the ER or Inpatient. This suggested that we can expect missing data problem for this outcome in data sources that do not have inpatient/er data capture causing sensitivity errors.
  • Characterization: top covariates by domain on day 0 were
    Condition Occurrence: weakness of facial muscles, type 2 diabetes mellitus, headache, altered sensation of skin, anesthesia of skin, polyneuropathy, Tear film insufficiency
    Procedure occurrence: 10%-20% had CT head , MRI use around 10%

Potential sources of error:

  • I did not observe significant evidence from cohort diagnostics to support sensitivity and index date misclassification errors. I found some evidence of specificity errors such as observed some Cerebral infarction due to thrombosis of cerebral arteries, Acute ill-defined cerebrovascular disease, Atrial fibrillation, and Transient cerebral ischemia on day 0. This may suggest that there may still be persons with UMN (stroke) as the cause of facial palsy.

Overall: this cohort definition appears to represent the target clinical idea well.

Performance characteristics

Pending - PheValuator

see https://data.ohdsi.org/PhenotypeLibrary/ cohort id
C256: [P] Facial Palsy lower motor neuron including Bells Palsy (180Pe, 0Era, 183W)

@Evan_Minty could you please review

Yes, will take this on.