Epidemiology

The incidence of Bell palsy varies between 20 and 30 per 100,000 people per year3,4; it accounts for ∼60–75% of all acute peripheral facial palsy.5 The male/female ratio for Bell palsy is approximately equal. The median age at onset is 40 years, but the disease may occur at any age.6 The incidence is lowest in children under 10 years old and increases from the ages of 10 to 59 ( Fig. 9.1 ). The left and right sides of the face are equally involved. There is no seasonal variation.7 Bell palsy is occasionally complicated by diabetes mellitus or hypertension. In a report on 625 patients with Bell palsy, diabetes mellitus was associated in 7% and hypertension in 14.1%, respectively.8 In another report, diabetes mellitus associated in 4.2–6.6% of Bell palsy, which is higher than the rate of 1.7–5.4% in the general population.9

Peitersen reported that 30% of patients had incomplete paralysis and 70% had complete paralysis; however, in 922 of our patients with Bell palsy, 5.4% had House-Brackmann grade II, 12.4% grade III, 34.4% grade IV, 26.3% grade V, and 21.7% grade VI ( Fig. 9.2 ).7

Approximately 70% of patients with Bell palsy recover completely without treatment, but 20–30% may have permanent disfiguring facial weakness or other permanent sequelae, such as synkinesia and contracture, when medical treatment fails.7 Compared with adults, the prognosis for facial nerve paralysis is generally favorable in children.7,10 Poor prognostic factors are older age,3 complete facial palsy, pain in the ear,11 diabetes mellitus,11 hyper-tension,11 and/or impaired taste.12

Recurrence of Bell palsy occurs in 7.1–12% of patients,13,14 with ipsilateral recurrence and contralateral involvement being roughly equal.13 Patients with recurrences are more likely to have a family history of Bell palsy.15 The incidence of diabetes mellitus in patients with recurrent Bell palsy is reported to be 2.5 times higher than that noted in nonrecurrent cases.16

Etiopathogenesis

Many hypotheses have been proposed for the cause of Bell palsy, including viral infection,17 microcirculatory failure of the vasa nervorum,18,19 ischemic neuropathy, and auto-immune reactions.20,21 Of them, the viral inflammatory immune concept has gained the most support from clinical observations and experimental studies reported over the past 30 years.

Viral Etiology

Acute facial paralysis can occur as part of many viral diseases, such as infectious mononucleosis caused by the Epstein-Barr virus,22 labial herpes (HSV),23 chickenpox (VZV),24 poliomyelitis (poliovirus),25 mumps,26 rubella,27 adult T cell lymphoma,28 and acquired immunodeficiency syndrome (human immunodeficiency virus).29 Each viral disease has a unique presentation, but in many ways the facial paralysis is similar to that of Bell palsy: transient and with overall good outcome. This suggests that viral infection may cause facial nerve paralysis; however, patients with Bell palsy rarely have unique viral presentation other than facial nerve paralysis. Common cold sore virus HSV-1 is characterized by recurrent vesicular eruptions of the oral mucosa or lips. Primary HSV-1 infections usually occur in early childhood and are frequently asymptomatic, or may present as gingivostomatitis. The virus also has infectious affinity to the nerves, with latent infection in the neuronal cells and neuropathogenicity. These features of HSV-1 have suggested it as the most probable cause of Bell palsy.

In 1972, McCormick hypothesized that HSV is the cause of Bell palsy.17 He suggested that HSV might be present in the geniculate ganglion of the facial nerve, where reactivation could cause a facial nerve neuropathy and infect the Schwann cells. Since then, many clinical and experimental studies have been attempted to obtain evidence for the HSV hypothesis. Serological studies were conducted by Adour et al30 and Vahlne et al.31 Using a complement fixation (CF) test, they demonstrated a higher prevalence of HSV antibodies in patients with Bell palsy than in the general population, suggesting previous exposure to HSV. However, they failed to find significant change in HSV antibody titers, or seroconversion, from the acute to convalescent phase, which would provide evidence for viral causality. This result is explained by the fact that reactivation of HSV-1, unlike VZV, can occur without a measurable antibody response. Nakamura et al32 analyzed more sensitive and specific neutralization antibodies for HSV-1 and found 15% of patients with Bell palsy to be positive for these antibodies. Serological examination with CF and neutralization antibodies is an indirect test and does not provide direct evidence for the cause of Bell palsy.

Mulkens et al33 first suggested a direct link between Bell palsy and HSV infection by cultivating HSV-1 from facial nerve epineurium obtained during decompression surgery. They isolated HSV from the specimens of one of two patients. On the other hand, Palva et al34 failed to isolate the virus from cultivated neural tissue. Virus cultivation is specific but insensitive and inconsistent for identification of HSV. Murakami et al35 used the powerful tool of polymerase chain reaction (PCR) to identify the HSV-1 and VZV genomes. They analyzed specimens of endoneurial fluid and postauricular muscle obtained during decompression surgery from patients with Bell palsy, Ramsay Hunt syndrome, and controls. The HSV-1 genome was detected specifically in 11 of 14 (79%) patients with Bell palsy, whereas the VZV genome was detected specifically in 8 of 9 (89%) patients with Ramsay Hunt syndrome. All controls tested negative for the HSV-1 and VZV genome. HSV-1 and VZV usually remain dormant in the geniculate ganglion of the facial nerve36,37 and would probably not be detected in the endoneurial fluid or auricular muscle unless they were reactivated. Therefore, Murakami et al′s study showed a direct association between Bell palsy and HSV-1 as well as Ramsay Hunt syndrome and VZV. Burgess et al38 also amplified the HSV genome from the geniculate ganglion on the affected side of a patient who died 6 days after developing Bell palsy, lending further support to HSV etiology.

Animal experiments also support the HSV etiology of Bell palsy. Kumagami39 made the first attempt to create an animal model of facial nerve paralysis using HSV. He injected HSV directly into the facial nerve at the stylomastoid foramen of the rabbit. He succeeded in causing facial nerve paralysis in 16 of the 19 animals within 6 days of virus inoculation. However, no animals except one showed improvement of the facial nerve paralysis in 223 days of follow-up. Thomander et al40 inoculated two different neuropathogenetic strains of HSV-1 (KJ 502, F) into the tongues of mice and demonstrated virus antigens in various brainstem areas, including the facial motor nucleus, but they failed to create a model of facial nerve paralysis. Ishii et al41,42 inoculated the Tomioka strain of HSV-1 into nasal mucosa, tongue, oral muscles, auricle, and intratemporal facial nerve of the mouse. They succeeded in inducing facial nerve paralysis only by inoculating the virus directly into the intratemporal facial nerve; inoculating the virus into the nasal mucosa, tongue, oral muscles, or auricle failed to produce paralysis. None of these attempts were successful in producing acute and transient facial paralysis resembling Bell palsy.

In 1995, Sugita et al43 first succeeded in producing acute and transient facial paralysis in 57% of mice by inoculating the KOS strain of HSV-1 into the auricle of 4-week-old female mice. Facial nerve paralysis developed 6 to 9 days after virus inoculation, after which there was spontaneous recovery within 14 days. Histopathological findings of the facial nerve in this animal model resembled those seen in reports of patients with Bell palsy, with nerve swelling ( Fig. 9.3 ), inflammatory cell infiltrates, and vacuolar degeneration.43,44

Honda et al45 clarified the pathophysiology underlying facial nerve paralysis using electrical tests of the blink reflex and electroneurography with histopathological examinations. They confirmed that the basis of HSV-1 neuritis was mixed lesions of various nerve injuries ( Fig. 9.4 ) and that recovery on electroneurography tended to be delayed compared with the recovery of the facial nerve paralysis. Murakami46 and Hato47 clarified the role of HSV-1 infection and immune functions in the pathogenesis of mouse facial nerve paralysis. They traced the migration route of the injected HSV from the auricle to the facial nerve and brainstem, and concluded that HSV infection in the facial nerve is a prerequisite for facial nerve paralysis. Hato′s immunological experiment suggested that facial nerve paralysis is caused by direct viral injury rather than viral-induced cellular demyelination.47 These murine models, however, are produced by primary infection with HSV-1, whereas Bell palsy is thought to be caused by reactivation of latent HSV-1 present in the geniculate ganglion. In 2001, Takahashi et al48 developed the HSV-1 primary infection model into a reactivation model. They suppressed cell-mediated immunity using anti–cluster of differentiation 3 and then, 8 weeks after recovery from facial nerve paralysis, scratched auricular skin with a needle in the same area as the previous HSV-1 inoculation. In this experiment, facial nerve paralysis developed in 20% of the mice, whereas the HSV-1 genome was detected in 67% of facial nerves. This study confirmed that reactivation of HSV-1 can produce facial nerve paralysis similar to that in Bell palsy and also suggested that reactivation of HSV-1 in the geniculate ganglion does not always cause facial nerve paralysis, so-called asymptomatic or subclinical reactivation of HSV-1.

It is widely recognized that HSV in the trigeminal ganglion can produce dozens of episodes of labial herpes; thus, one might wonder why, if HSV causes Bell palsy, the episodes are rarely repeated. The study by Takahashi et al48 suggested an answer to this question by showing that facial nerve paralysis does not always appear even if HSV-1 is reactivated in the geniculate ganglion of the facial nerve. The manifestations and frequency of episodes developing in sensory and motor nerves seem somewhat different. Other factors, such as the anatomical structure of the facial canal, may be closely related to the pathomechanism of facial nerve paralysis.

Ischemia

Blood supply to the facial nerves derives from three arteries: the labyrinthine artery, middle meningeal artery, and stylomastoid artery. Microcirculatory failure of the vasa nervorum,18,19 or ischemic neuropathy, is the most traditional hypothesis for the cause of Bell palsy. In 1944, Denny-Brown and Brenner49 demonstrated that the demyelination and disruption of nerve fibers with loss of nerve conductivity was caused by ischemic injury from compression of arterial branches supplying the nerve, rather than by direct compression of axoplasmic flow. Hilger18 postulated that Bell palsy is an ischemic neuritis resulting from segmental arteriolar spasm, based on clinical observations and Denny-Brown and Brenner′s experimental study. Calcaterra et al50 reported two cases of total unilateral facial paralysis after embolization of the middle meningeal artery, which suggested that ischemia of the tympanic portion of the facial nerve might be responsible for facial nerve paralysis. Kumoi et al51 developed an animal model of ischemic facial palsy by embolization of internal and external maxillary arteries in cats. In this animal model, facial nerve paralysis developed immediately after embolization and recovered spontaneously after 2 months.

Autoimmune Injury

Autoimmune reaction against myelin components is a possible cause of peripheral nerve demyelination such as Guillan-Barré syndrome and multiple sclerosis. Abramsky et al20 found that peripheral blood lymphocytes in patients with Bell palsy showed significant transformation in the presence of human basic protein from peripheral nerve myelin. McGovern et al21 found that more profound facial nerve paralysis and severe histologic nerve damage developed in hyperimmune dogs than in nonsensitized dogs after injection of vasoconstrictors into the facial nerve. In further study, they found that cromolyn sodium, a mast cell degranulation inhibitor, prevented the neuropathic changes induced by horse serum injection into the perineurial space of the facial nerve. They concluded that degranulation of mast cells activated by complement or specific allergens may be the triggering mechanism that leads to nerve edema, ischemia, and paralysis.

Evidence for cellular and humoral autoimmune mechanisms of nerve injury has been reported in patients with Bell palsy. Jonsson et al52 found an increase in levels of interferon-γ, but failed to find any change in interferon in the acute versus convalescent phases of facial nerve paralysis. They thought this increase of interferon may be representative of a chronic viral infection or reactivation. Yilmaz et al53 examined serum cytokines in 23 patients with Bell palsy and found significantly higher levels of interleukin-6, interleukin-8, and tumor necrosis factor-α than in controls. They suggested that high serum interleukin-6 indicated enhanced repair of injury in astrocytes, and that high serum tumor necrosis factor-α level may have led to replication of HSV and an inflammatory process of virus-induced demyelination. In contrast, Bujía et al54 found no difference in serum levels of soluble interleukin-2 receptors, which reflect T-lymphocyte activation, in patients with Bell palsy and age- and sex-matched controls. They concluded that T-cell activation was not a prominent feature in Bell palsy. Taking all of this into consideration, while immunological reaction may have important role in the pathogenesis of facial nerve paralysis, it is unlikely to be the primary cause of Bell palsy because autoimmune reactions occur regardless of whether they are triggered by antigens of virus, bacteria, parasites, or other foreign substances.