Many events, such as viral infection [1, 3, 4], ischemia [5], and autoimmune reaction [6], have been proposed as causes of Bell palsy. Viral infection is thought to be the most likely cause [7]. However, it is rare to find a diagnostic fourfold increase in specific viral antibody titer in the acute and convalescent serum specimens of patients with Bell palsy [3, 4, 7]. Postmortem histopathologic studies of the facial nerve suggest viral neuritis [8], but electron microscopic studies have failed to detect specific viral particles in the facial nerve [9]. Because the etiologic agent of Bell palsy is unknown, treatment of this condition is empiric, varying from observation alone to the use of steroids, surgical decompression, and antiviral agents.

We analyzed the viral genomes of herpes simplex virus type 1 (HSV-1), varicella-zoster virus, and Epstein-Barr virus using polymerase chain reaction (PCR) on facial nerve endoneurial fluid specimens and specimens of posterior auricular muscle innervated by the facial nerve.

Methods

Top Methods Results Discussion Author & Article Info References

Patients and Specimens

During a 4-year period, 14 of 170 patients with Bell palsy and 9 of 51 patients with the Ramsay-Hunt syndrome had decompression surgery 12 to 87 days after the onset of facial palsy. None had benefited from medical management. All patients and controls gave informed consent.

Two types of clinical specimens were collected intraoperatively: endoneurial fluid from the facial nerve and tissue from the posterior auricular muscle. A piece of auricular muscle was resected after skin incision, and we obtained endoneurial fluid by absorbing it with a small, sterilized surgical sponge held at the mastoid segment immediately after the epineural sheath was incised. We stored both specimens immediately at – 80 °C and continued to store them at that temperature until PCR analysis was done. Control specimens of endoneurial fluid and posterior auricular muscle were collected during decompression surgery from four patients with temporal bone fracture or bacterial infection concomitant with otitis media. Posterior auricular muscle specimens were obtained during tympanoplasty from five patients with chronic otitis media who did not have facial paralysis. As an additional control, a piece of neural tissue was obtained from each of three patients with parotid tumors or facial neuroma whose facial nerves had already been affected (Table 1).

Polymerase Chain Reaction

To amplify and identify the HSV, varicella-zoster virus, and Epstein-Barr virus genomes, five sets of virus-specific primers and internal oligonucleotide probes were synthesized for PCR and Southern blot analysis. Primer set 1 was prepared for amplification of the HSV-1 genome, which is located on the US6 gene [10]. Primer set 2 was designed to amplify both HSV-1 and HSV-2. A sense primer (5'-CCACCGAGCGGCAGGTGATC-3') and an antisense primer (5'-GCCGACCGCCTGCTCGTGCT-3') are located on the UL44-45 gene [11]. Using primer set 2, we discriminated between HSV-1 (578 base pairs) and HSV-2 (621 base pairs) on the basis of size. Nucleotide sequences of the HSV specific internal probe are 5'-GAGGCGATCGAGTGGGT-3'. Primer sets 3 and 4 were prepared for varicella-zoster virus amplification (genes 29 and 62, respectively) [12], and primer set 5 was prepared for Epstein-Barr virus amplification (latent cycle gene) [13].

The sensitivities of the primer sets were assessed by making serial 10-fold dilutions of each purified DNA sample. The limits of detection for primer sets 1, 2, 3, 4, and 5 were about 10, 100, 10, 100, and 10 femtograms, respectively.

Samples of endoneurial fluid (about 10 µL), posterior auricular muscle (2 mg to 5 mg), and nerve tissue (0.5 mg to 1 mg) were completely digested with proteinase K. Polymerase chain reaction amplification and subsequent hybridization with Southern blot analysis were done as described previously [14]. Rigid precautions against contamination in the sample processing included the use of water controls replacing the DNA samples in all amplifications.

Serum Antibody Titers

We examined serum antibody titers by using the complement fixation test for HSV-1 and varicella-zoster virus and by using fluorescent antibody methods for Epstein-Barr virus 8 to 38 days after the onset of facial paralysis.

Results

Top Methods Results Discussion Author & Article Info References

We amplified HSV-1-specific DNA fragments from the US6 and UL44-45 genes in both endoneurial fluid and posterior auricular muscle specimens obtained from patients with Bell palsy. The PCR-amplified products of the US6 gene were detected by Southern blot analysis in 10 of the 13 fluid specimens (77%) and 8 of the 14 muscle specimens (57%); the products of the UL44-45 gene were detected in 4 of the 13 fluid specimens (31%) and 7 of the 14 muscle specimens (50%). Neither varicella-zoster virus nor Epstein-Barr virus was detected in the same clinical specimens (Figure 1, top; (Table 1). We did not detect HSV-1 DNA in either the fluid or the muscle specimens of three patients with Bell palsy [patients 9, 12, and 13]. The PCR-amplified DNA fragments of the US6 gene from two patients with Bell palsy (patients 6 and 7) were sequenced directly after asymmetric PCR was done as described previously [14]. The nucleotide sequences of the amplified products (221 base pairs) were identical to those of the HSV-1 genome submitted to the GenBank (Mountain View, California [data bank with genetic information]) with accession numbers J02217 and K02372.

View larger version (94K): [in this window] [in a new window]   Figure 1. Amplification of herpes simplex virus type 1 (HSV-1) and varicella-zoster virus (VZV) genomes from clinical samples. Ethidium bromide staining (top) and Southern blot analysis (bottom) of polymerase chain reaction products obtained from endoneurial fluids (lanes 1 to 9, top; lanes 1 to 9, middle; and lanes 1 to 3, bottom), nerve tissues (lanes 4 to 6, bottom), and posterior auricle muscles (lanes 7 to 9, bottom) are shown. Using the primer set 1, HSV-1 specific 221-base pair (bp) DNA fragments were amplified only from patients with Bell palsy (top). On the other hand, 272-base pair fragments specific to varicella-zoster virus were amplified only from patients with the Ramsay-Hunt syndrome by the primer set 3 (middle). Plasmid DNAs used as a carrier for DNA extraction are seen in lanes 1 to 9 but not in lanes M and P. M equals molecular size marker of 100-base pair ladder; P equals positive control amplified from each viral genome. Lanes 1 to 9 correspond to patients 1 to 9.

We detected varicella-zoster virus DNA from gene 29 or gene 62 in specimens obtained from patients with the Ramsay-Hunt syndrome only Figure 1, middle; (Table 1). Gene 29 was detected in 8 of the 9 patients (89%), and gene 62 was detected in 6 of the 9 patients (67%). On the other hand, we could not amplify HSV-1, varicella-zoster virus, or Epstein-Barr virus DNA from any specimens obtained from the other controls Figure 1, bottom; (Table 1).

Serum antibody titer to HSV-1 was positive in 12 of 13 patients (92%) with Bell palsy, in 4 of 9 patients (44%) with the Ramsay-Hunt syndrome, and in 5 of 9 controls (56%) (Table 1). The prevalence of HSV-1 antibody in patients with Bell palsy was significantly higher than that in controls (P < 0.05, Fisher exact test). However, antibody titers to HSV-1 in patients with Bell palsy were not significantly higher than those of controls, as previously reported [3, 4].

Discussion

Top Methods Results Discussion Author & Article Info References

We found HSV-1 DNA in 11 of 14 patients (79%) with Bell palsy, and we found varicella-zoster virus DNA in 8 of the 9 patients (89%) with the Ramsay-Hunt syndrome. The identification of viral DNA may not always be definitive evidence that a particular agent causes a disease process, because PCR can amplify viral DNA regardless of whether the virus is in the infective, lytic, or latent state. The presence of latent HSV-1 and varicella-zoster virus genomes has also been shown by PCR in the geniculate ganglion of human facial nerves at autopsy [15-17]. However, HSV-1 and varicella-zoster virus usually remain dormant in ganglia and would probably not be detected in the endoneurial fluid or auricular muscle unless they were reactivated. This hypothesis was supported by our inability to detect either HSV-1 or varicella-zoster DNA in controls. Cranial nerve surgery often reactivates latent HSV, causing labial and facial herpetic lesions 48 to 72 hours after surgery [18]. However, because we obtained specimens within 2 hours of beginning decompression surgery, reactivation of the virus in the muscle or fluid was probably not induced by surgery. If this surgery did reactivate latent HSV-1, viral DNA should also have been detected in patients with the Ramsay-Hunt syndrome and in other controls who were seropositive for HSV-1.

Triggers known to be associated with Bell palsy are also known to reactivate HSV. Preceding stress, such as upper respiratory tract infection, fever, dental extraction, menstruation, or exposure to cold might reactivate latent HSV-1 in the geniculate ganglion. After the virus reactivates, it destroys ganglion cells and spreads into the endoneurial fluid. The virus also infects Schwann cells, leading to demyelinization and inflammation of the facial nerve [19]. This inflammatory response has been shown by gadolinium-enhanced magnetic resonance imaging in patients with Bell palsy and in patients with the Ramsay-Hunt syndrome [20].

Given the known neuropathogenicity of HSV-1 and the presence of HSV-1 DNA in the lesional site of the facial nerve specific to patients with Bell palsy, we conclude that HSV-1 infection in the facial nerve is directly related to the pathogenesis of Bell palsy just as the varicella-zoster virus is directly related to the pathogenesis of the Ramsay-Hunt syndrome. There are two possible explanations for our failure to detect HSV-1 in three of the patients with Bell palsy [patients 9, 12, and 13]: 1) the limited sensitivity of PCR analysis to detect small amounts of viral DNA and 2) the presence of an etiologic agent other than HSV-1. More data are required to determine the percentage of patients with Bell palsy in whom HSV-1 is the etiologic agent of Bell palsy, but our findings suggest that HSV-1 infection is the major cause of Bell palsy and that treatment with appropriate antiviral agents might benefit most patients with this condition.

Drs. Mizobuchi, Nakashiro, and Doi: Department of Neuropsychiatry, Ehime University School of Medicine, Shigenobu-cho, Onsen-gun, Ehime 791-02, Japan.