Approach to new-onset facial nerve palsy in a critically ill patient: A case report.

Lower motor type of facial nerve palsy (Bell's palsy) is one of the most common types of facial nerve palsy, with well-defined management with steroids and antivirals for patients attending outpatient clinics. The diagnosis and management of facial nerve palsy in critically ill patients require an individualized approach, as there may be many other causes like soft-tissue compression due to facial edema which can occur as a complication of prone ventilation and severe subcutaneous emphysema. This report highlights the challenges in the management of new-onset facial nerve palsy diagnosed in the intensive care unit (ICU) for a patient on mechanical ventilation, and creates a necessity for a standard operating protocol for the management of such scenarios in ICU. Copyright:

INTRODUCTION

Facial nerve palsy has various etiologies such as cerebrovascular accident, infections (herpes, Lyme disease, and otitis media), compression of surrounding skin and soft tissue due to trauma (fracture of temporal bone), tumors (parotid gland), demyelinating disorders such as Guillain–Barre syndrome, and systemic disorders such as sarcoidosis and multiple sclerosis.[12] Bell's palsy is one of the most common types of facial nerve palsy, which has streamlined management approach with steroids and antivirals, if the patient presents early to outpatient clinic.[2] Through this report, we want to highlight the challenges in the management of new-onset facial nerve palsy diagnosed in the intensive care unit (ICU) for a patient on mechanical ventilation, and thereby create a necessity for a standard operating protocol for the management of such scenarios in ICU.

CASE REPORT

A 32-year-old female with 34 completed weeks of gestation and no comorbidities was admitted to the ICU with severe coronavirus disease 2019 (COVID-19). On admission, she was in hypoxemic respiratory failure (SpO2: 94% with oxygen therapy by reservoir bag at 15 L/min and respiratory rate of 30 breaths per min). She was treated with piperacillin–tazobactam [4.5 g intravenous (IV) every 6 h for 7 days], methylprednisolone dose of 1 mg/kg in divided doses for 5 days, remdesivir (200 mg IV on day 1 followed by 100 mg IV once daily for the next 4 days), and other ICU care, along with high-flow nasal cannula therapy (flow rate at 60 L/min; FiO2:1) for 6 h. She was intubated for persistent hypoxemic respiratory failure, and standard lung-protective ventilation was initiated. Pregnancy was terminated, as acute respiratory distress syndrome was worsening and nonstress test revealed fetal distress. A lower segment cesarean section (LSCS) was performed, and a 2-kg live female baby was delivered. The patient's PaO2/FiO2 ratio (P/F ratio) post LSCS was 120, and she was given two prone ventilation sessions (duration of 18 h each). There was a significant improvement in her oxygenation (P/F Ratio to 300) by the end of second prone session.

Upon supination, there was diffuse subcutaneous edema of the face, which reduced progressively over the next 4 days. Since prolonged mechanical ventilation was anticipated, percutaneous tracheostomy was done on day 7 of ICU stay. Later, her ICU course was complicated by septic shock with Klebsiella pneumoniae bacteremia, which was treated with antimicrobials as per sensitivity report (colistin and meropenem). The shock recovered in the next week and weaning was initiated with a spontaneous breathing trial on the 16th day of ICU stay, during which it was noticed that the patient had inability to close her left eyelid and there was deviation of the angle of mouth to the right side. There was no other focal neurological deficit, and plantar reflexes were of flexor response. Examination of both the ears did not reveal any discharge. The power of all four limbs was reduced with a Medical Research Council sum score of 40, with diagnosis of critical illness neuromyopathy (CINM). Patching of the eyelid and lubrication with eye drops were done to prevent exposure keratopathy. A contrast-enhanced computed tomography scan for the head, face, and temporal region was done to rule out any trauma due to prone ventilation session. It revealed soft-tissue swelling with fat stranding in the left cheek and masseter muscle, with no evidence of stroke (bleed/infarct), no significant compression in the tract of the facial nerve, and no fracture or any osseous lesion in the facial canal [Figure 1a]. CINM was managed with active limb physiotherapy and good glycemic control. Steroids were deferred in view of CINM and her recent recovery from septicemia, apart from lack of clear onset time. Acyclovir was not administered as inflammation secondary to compression of soft tissue due to prone positioning is the suspected etiology (not infection), and there is limited evidence to support its use without concomitant steroids.

Figure 1

(a) Computed tomography scan of the base of the skull (bone window) revealing normal facial canal (shown in red arrow) with no evidence of any fracture or osseous lesion noted on either side; (b) Magnetic resonance T1 postcontrast images showing subtle enhancement in the cisternal and intracanalicular segment of the left facial nerve (shown in red arrow)

As there was an improvement in her neuromuscular strength, she was gradually weaned from mechanical ventilation by the 23rd day of ICU stay and decannulated by the 30th day. The left facial nerve palsy persisted with House–Brackmann Facial Nerve Grade IV. As we could not explain the etiology of facial nerve palsy, magnetic resonance imaging of the brain with contrast was done which revealed normal study other than subtle postcontrast enhancement in the cisternal and intracanalicular part of the 7th and 8th nerve complex in T1 + C-weighted image [Figure 1b]. The patient was discharged home after 1 month of ICU stay with instructions for eye care. On follow-up 2 months following discharge from ICU, there was complete resolution of paresis without any serious long-term effects like crocodile tears syndrome.

DISCUSSION

Through this case, we want to mention that the management of facial nerve palsy in outpatient clinic is completely different to its management in critically ill patients, due to various conglomerating factors. The existing literature suggests that the administration of steroids can lead to good outcomes if therapy was given within 3 days of symptom onset, but no such study included critically ill patients on mechanical ventilation.[23] With the known adverse effects of steroids such as hyperglycemia, peripheral edema, exacerbation of myopathy, infections, and reactivation of stress ulcer, steroids need to be used very carefully in critically ill patients. The recovery from facial nerve palsy without any pharmacological therapy is seen in around 70% of patients within 6 months, and up to 80% within 9 months.[34] Citing the abovementioned reasons (unknown onset: >72 h, recent recovery from sepsis, CINM, and anticipated recovery without any pharmacological treatment in up to 80% of patients), the index case was not treated with steroids. As inflammation is the suspected etiology, and as there was not much evidence for administration of antivirals alone without steroids, they were deferred.[245] Inflammation and edema causing compression to the facial nerve passing through the fallopian canal in temporal bone is the most common mechanism of Bell's palsy.[1] The plausible hypothesis of facial nerve palsy in our case may be compression of soft tissue due to prone positioning exacerbated by postpregnancy physiological status or late COVID-19 sequelae, both of which are known risk factors for facial nerve palsy.[6] This was supported by imaging done. Prone ventilation has been reported to causelower cranial nerve palsy which occurs due to compression from edema to cranial nerves IX to XII travel which travel together through the mandibular retrocondylar-peripharyngeal space.[6] Our case could have had such palsy due to edema in the region of the fallopian canal, compressing the facial nerve. Facial edema following prone positioning can be prevented through frequent head rotation, head end elevation, and prevention of compression through extra padding or cushioning.[6]

The purpose of reporting this case was to highlight the knowledge gap and propose an algorithm for diagnostic workup and management of new-onset facial nerve palsy in the ICU patients, keeping various other etiologies in workup [Figure 2].[78910] Patients with incomplete recovery after 3 months, or with new/worsening ocular/neurological symptoms, should be re-evaluated for other causes such as head and neck/cutaneous malignancy. Care of such patients should be focused on prevention of long-term functional impairment addressing incomplete eye closure, lacrimal dysfunction, motor synkinesis, or crocodile tears syndrome, nasal airway obstruction, and oral incompetence.[2]

Figure 2

Approach to management of new-onset facial nerve palsy in the intensive care unit. #Last neurological examination done with in the past 48 h within normal limits; *After risk–benefit assessment, could be considered if acute; R/O: Rule out, ENT: Ear, nose, and throat, CVA: Cerebrovascular accident, H/O: History of, CSF: Cerebrospinal fluid, VZV: Varicella-zoster virus, HSV: Herpes simplex virus, CECT: Contrast-enhanced computerized tomography, CINM: Critical illness neuromyopathy

CONCLUSION

Diagnosis and management of facial nerve palsy in critically ill patients require an individualized approach. There is a need to create recommendations for the management of such conditions in critically ill patients on mechanical ventilation.

Research quality and ethics statement

This case report did not require approval by the Institutional Review Board/Ethics Committee. We followed applicable EQUATOR Network (http://www.equator-network.org/) guidelines, specifically the CARE guideline, during the conduct of this research project.

Declaration of patient consent

Written informed consent was provided to publish de-identified case details by the patient's surrogate.

Financial support and sponsorship

None.

Conflicts of interest

There are no conflicts of interest.