Introduction
Facial nerve (FN) palsy is a potentially devastating complication of otological surgery, resulting in weakness of ipsilateral muscles of facial function and subsequent functional, aesthetic, emotional and medico-legal implications.Reference Bae, Kwak, Nam and Jung1, Reference Hotton, Huggons, Hamlet, Shore, Johnson and Norris2 Post-operative palsy can broadly be classified into immediate and delayed, depending on the onset time, the definitions of which vary in the literature.Reference Bae, Kwak, Nam and Jung1, Reference Patel, Walters, Kim and Walijee3 The House–Brackmann (HB) grading system is a widely used system to quantify FN function on a scale of I (normal) to VI (total paralysis). Other facial palsy classification systems, such as the Sunnybrook Facial grading system, are also widely used.Reference Mat Lazim, Ismail, Abdul Halim, Nik Othman and Haron4
For delayed facial palsy (DFP), the incidence varies from 0.2 to 1.9 per cent for middle ear surgeries and 0.64 per cent for cochlear implantation.Reference Bae, Kwak, Nam and Jung1, Reference Eckermann, Meyer and Guenzel5 The most common theory behind DFP is the reactivation of a latent herpes simplex virus (HSV) in the geniculate ganglion. Other theories include intra-operative neuronal oedema causing ischaemia to an enclosed FN canal, direct pressure on the tympanic segment of the FN or vasospasm of feeding vessels.Reference Bae, Kwak, Nam and Jung1, Reference Lovin, Sweeney, Chapel, Alfonso, Govil and Liu6, Reference Bonkowsky, Kochanowski, Strutz, Pere, Hosemann and Arnold7 At present, there is no evidence-based therapy for DFP, with most studies using corticosteroids as mono- or combination therapy with anti-viral agents largely based on treatment guidance for idiopathic Bell’s palsy.Reference Eckermann, Meyer and Guenzel5
This systematic review aimed to identify and synthesise documented cases of DFP following otological surgery and to assess management options and outcomes.
Materials and methods
Study registration
This study was registered on the PROSPERO database of systematic reviews (reference CRD42022279684). The 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed.Reference Page, McKenzie, Bossuyt, Boutron, Hoffmann and Mulrow8
Inclusion and exclusion criteria
Studies for inclusion were clinical studies of FN palsies following middle ear and cochlear implantation surgery in adults and children. Studies included were experiential or observational of human subjects. Animal or laboratory studies and cases where the aetiology of DFP was traumatic were excluded, as were studies focusing on surgical techniques and those focusing on neuro-otological and lateral skull base procedures.
Study search strategy
MEDLINE, Pubmed, Embase and The Cochrane Central Register of Controlled Trials were searched by a clinical librarian. The search strategy is shown in Supplementary Figure 1. There were no restrictions on language or date.
Study selection
Two reviewers (S.P./H.W.) independently screened titles and abstracts. Full texts of studies potentially meeting the inclusion criteria were then reviewed by three reviewers (S.P./H.W./V.K.) to assess eligibility. Authors were contacted directly when full texts could not be found or there was insufficient data. If studies remained unavailable, they were excluded. Conflicts between reviewers were resolved by discussion. Reference lists of screened full texts were examined and reviewed for potential additional studies which met the inclusion criteria. The primary outcome measure was the incidence of DFP (permanent/temporary), management modalities and outcomes (HB FN grading pre- and post-treatment for studies published after its introduction). Secondary outcomes were the use of intra-operative nerve monitoring, immune titres and time to recovery.
Data extraction
Data were independently extracted by using a pre-determined spreadsheet (Microsoft Excel).
Risk of bias analysis
The JBI critical appraisal tools were used to assess the risk of bias, and the Oxford Centre for Evidence-Based Medicine (OCEBM) grading system was used to assess the level of evidence of each article.Reference Aromataris, Lockwood, Porritt, Pilla and Jordan9, 10 Biases were independently assessed by S.P. and B.W. using a spreadsheet (Microsoft Excel), with conflicts resolved by discussion.
Results
Search results
Searches were performed on 10 May 2024, and Figure 1 shows the results of screening. The initial search identified 132 abstracts, of which 51 were shortlisted for full-text review. Thirty-nine articles were retrieved from reference lists. Thirty-five did not meet the inclusion criteria and were excluded, leaving 49 articles for qualitative analysis.Reference Lovin, Sweeney, Chapel, Alfonso, Govil and Liu6, Reference Bonkowsky, Kochanowski, Strutz, Pere, Hosemann and Arnold7, Reference Ahn, Chung and Lee11–Reference Zohar and Laurian57 Study design and the inconsistent reporting of outcomes precluded quantitative analysis.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.
Characteristics of included studies
Study characteristics and outcomes are displayed in Table 1 and Supplementary Table 1, respectively. This review included 33 case series, 11 case reports, three retrospective cohorts and two case-control studies from 1967 to 2021. Of 24,917 operations, there were 201 episodes of DFP. The mean onset of DFP was post-operative day eight. The grade of palsy ranged from II-VI. Thirteen studies analysed viral titres: herpes simplex virus (HSV) immunoglobulin (Ig) G and IgM and varicella zoster virus (VZV) IgG and IgM. Where reported, 93 per cent of DFPs (162/174) fully recovered. Where reported, time to complete recovery took less than or equal to seven days in nine cases, eight days to one month in 56 cases, one to three months in 31 cases, and more than three months in six cases.
Study characteristics
Abbreviations: DFP = Delayed facial palsy; OCEBM = Oxford Centre for Evidence-Based Medicine.
Tympanoplasty procedures
Twenty-seven cases of DFP following 1,945 tympanoplasty procedures were reported in four studies (Table 1). DFP was reportedly a result of viral reactivation, oedema or compression from an ear pack on a dehiscent FN. Li et al. found no significant difference between the development of DFP in patients undergoing endoscopic ear surgery when compared with microscopic ear surgery for Type I tympanoplasty.Reference Li, Chan, Chen, Kuo, Wang and Chang33 Only Kalamden et al. reported using an FN monitor which did not alarm.Reference Kamalden, Yusof and Misron31 Viral studies were tested in three studies.Reference Gyo and Honda25, Reference Kamalden, Yusof and Misron31, Reference Li, Chan, Chen, Kuo, Wang and Chang33 Gyo and Honda reported an eight-fold increase in VZV antibody titres and particularly an increase in IgG to VZV for their patient with DFP.Reference Gyo and Honda25 Kamalden et al. reported one patient with DFP was positive for both IgM and IgG to VZV.Reference Kamalden, Yusof and Misron31 All patients with DFP (n = 8) were found to have IgG for HSV and VZV but not IgM for HSV and VZV in the study by Li et al., suggesting previous exposure but no recent infection.Reference Li, Chan, Chen, Kuo, Wang and Chang33 No studies reported a vesicular rash. Corticosteroids were administered to all patients in all studies. Combination therapy of corticosteroids with anti-viral therapy was administered to four patients by Li et al., all of whom recovered completely to HB grade I.Reference Li, Chan, Chen, Kuo, Wang and Chang33 The case by Gyo and HondaReference Gyo and Honda25 received corticosteroids, was decompressed for a “bulging” FN and recovered to HB grade III at 10 months.
Tympanomastoid operations
Thirty-eight cases of DFP following 2,610 tympanomastoid operations were reported in seven publications. The episodes occurred between 72 hours and 12 days post-operatively, with initial palsy severity between HB grades II and V. The authors attributed the aetiology of the DFP to viral reactivation (n = 3),Reference Basavaraj, Prakash, Shetty, Sandhya and Kallada15, Reference Safdar, Gendy, Hilal, Walshe and Burns43, Reference Vrabec53 oedema from thermal injury (n = 1),Reference Eskander, Holler and Papsin22 wound infection (n = 1),Reference Vrabec53 FN dehiscence (n = 9)Reference Xu, Liu, Zuo, Wang and Wang55 and transection of the chorda tympani (n = 1).Reference Thorne, Dunham and Tom52 Four studies reported VZV testing in those with DFP and found VZV IgM positive in 10 of 25 patients with DFP and IgG positive in nine out of 10 cases with DFP.Reference Xu, Liu, Zuo, Wang and Wang55 Treatment was with corticosteroid monotherapy or in combination with anti-viral therapy or antimicrobial therapy. In all studies other than that by Basavaraj et al., all patients returned to an HB grade I between one week and six months.Reference Basavaraj, Prakash, Shetty, Sandhya and Kallada15
Stapes operations
Forty-three cases of DFP were reported following 8,379 stapes operations over 13 studies (Table 1). The initial palsy severity ranged between HB grades II–VI and the time of onset ranged from post-operative days one to 41.
Where reported, the authors attributed the aetiology of the DFP to neuropraxia of unknown cause (n = 1),Reference Althaus and House13 viral reactivation (n = 3),Reference Ng and Maceri40, Reference Salvinelli, Casale, Vitaliana, Greco, Dianzani and D’Ascanio44, Reference Shea and Ge46 oedema from heat from a KTP laser (n = 1),Reference Mills, Szymanski and Abel38 Bell’s palsy of unknown origin (n = 1),Reference Zohar and Laurian57 FN dehiscence (n = 7);Reference Durcan, Shea and Sleeckx21, Reference Shea and Ge46, Reference Smith, Simon and Ramalingam48 and in one patient a dehiscent FN with aberrant anatomyReference Smith, Simon and Ramalingam48 and intra-operative transection of the chorda tympani.Reference Zohar and Laurian57
HSV antibody testing was performed in 14 cases. Salvinelli et al. reported IgM antibodies were positive in one of seven cases and IgG in six of seven cases.Reference Salvinelli, Casale, Vitaliana, Greco, Dianzani and D’Ascanio44 Shea and Ge reported antibody titres were positive in five of six cases.Reference Shea and Ge46 VZV antibody testing was performed in eight cases, with IgM and IgG positive in one, and unspecified antibody titres positive in all six by Shea and Ge.Reference Shea and Ge46 Salvinelli et al. reported the mean HSV immunoglobulin G titre was 1:14,050 in the subjects with DFP and 1:2,300 in controls; the difference was statistically significant (p < 0.001).Reference Salvinelli, Casale, Vitaliana, Greco, Dianzani and D’Ascanio44
Prophylactic acyclovir was used by both Salvinelli and Shea. Salvinelli et al. treated all patients with a history of herpes labialis with an acyclovir preventive therapy; a significant reduction in the prevalence of post-stapedectomy DFP was noted (7/114 vs 0/35; p < 0.001).Reference Salvinelli, Casale, Vitaliana, Greco, Dianzani and D’Ascanio44 Shea et al.Reference Shea and Ge46 used acyclovir prophylactically in a patient who had suffered from DFP from a previous stapedectomy and found this effective in preventing recurrence (n = 1).
Treatment consisted of corticosteroids (n = 8), anti-viral therapy (n = 7), combination corticosteroids and anti-virals (n = 4), corticosteroids and vasodilators (n = 2) or vasodilators and antibiotics (n = 1) and surgical decompression at four days following the onset of DFP (7 days post-operatively; n = 1). Where reported, the facial palsy resolved completely (n = 31), left a residual HB grade II palsy (n = 3) or left a “minimal paresis” (n = 1) at seven months in the patient who underwent decompression.
Cochlear implants
Eighty cases of DFP were reported following 8,966 cochlear implant surgeries across 22 studies. DFP onset ranged from 18 hours – 20 days post-operatively with one case by Alzhrani et al.Reference Alzhrani, Lenarz and Teschner14 reported after four years from the initial surgery. This was attributed to either chronic pressure from the electrode onto an exposed FN or an infection. Thermal propagation and resultant neuropraxia were proposed as a mechanism of DFP by four studies,Reference Black, Bailey, Albert, Leighton, Hartley and Chatrath17, Reference Farinetti, Ben Gharbia, Mancini, Roman, Nicollas and Triglia23, Reference Miyamoto, Young, Myres, Kessler, Wolfert and Kirk39, Reference Stratigouleas, Perry, King and Syms49 with viral activation suggested in two studies.Reference Ishikawa, Hosoya, Kanzaki and Ogawa28, Reference Joseph, Vishwakarma, Ramani and Aurora30 Other studies did not postulate an aetiology for the DFP. Fayad et al.Reference Fayad, Wanna, Micheletto and Parisier24 postulated that the DFP in a case of a 19-month-old was secondary to neural oedema resulting from an inflammatory response to the thermal injury of the proximity of the drill burr to the FN. Similarly, Miyamoto et al.Reference Miyamoto, Young, Myres, Kessler, Wolfert and Kirk39 reported that the FN was not exposed during surgery, and the authors suspected the DFP to be caused by heat generated by the burr shaft.
Two studies analysed viral titres. Joseph et al. found that their patient with DFP was positive for HSV IgM on polymerse chain reaction (PCR), with a 100 per cent increase in HSV titres, they also found a four-fold increase in VZV titres.Reference Joseph, Vishwakarma, Ramani and Aurora30 Ishikawa et al. identified an elevation in HSV antibody titres, although specifics were omitted.Reference Ishikawa, Hosoya, Kanzaki and Ogawa28
Where specified, patients were treated with either corticosteroids, anti-viral therapy, antimicrobial therapy or a combination of the three.
The majority (n = 71) of palsies recovered completely between three days and one year. There were cases of persistent grade II palsy (n = 1), grade III palsy (n = 1) and the outcomes were unspecified in the remainder of cases.
Unspecified operations
Bonkowsky et al.Reference Bonkowsky, Kochanowski, Strutz, Pere, Hosemann and Arnold7 reported outcomes following 1,800 unspecified middle ear operations. They observed seven ipsilateral DFP following five tympanoplasty and two stapes procedures. DFP onset was between 2.5 and 7 days post-operatively. In two cases, the chorda tympani was transected. Five patients with DFP were tested for HSV-1 using PCR, and four were positive. HSV-1 antibody titres were tested in seven patients. These were IgM positive in 1 and IgG positive in five cases. Data on recovery were provided for a single patient in the form of a case report of a patient who had an unspecified HB grade of palsy on the fifth post-operative day. This patient was commenced on anti-viral and corticosteroid therapy for 18 days. On the 11th day, there was an improvement in FN function to HB grade II, with complete recovery on day 34.
Viral titres
Thirteen studies reported viral immunoglobulin titres. HSV antibodies were tested in 31 cases (15 IgM, 15 IgG and one unspecified). HSV IgM antibodies indicative of an acute HSV infection were positive in one of 15 tested cases, and IgG antibodies, indicative of active or previous infection, were positive in 14 of 15 cases tested. This finding is non-specific as a positive IgG may indicate a previous infection. Unspecified HSV antibody testing was positive in 11 of 16 cases. VZV antibodies were tested in 50 cases (21 IgM, 20 IgG and 9 unspecified). VZV IgM antibody titres were positive in 10 of 21 cases, indicating an acute VZV infection. IgG VZV antibody titres were present in 18 of 20 cases. Unspecified VZV antibodies were positive in seven of nine cases.
Risk of bias in the included studies
The risk of bias assessment of each of the studies is summarised in Supplementary Tables 1–4. All studies included in this review were undertaken retrospectively with an OCEBM grade III–IV. Most studies did not disclose the qualifications, experience or facilities of authors however competence has been assumed. The cohort from which the case series were drawn was large enough to minimise selection bias. Only abstracts were available for the articles by Ng and MaceriReference Ng and Maceri40 and Vrabec,Reference Vrabec53 but the information available was suitable to include these within the review.
Discussion
Delayed FN palsy following otological surgery is not common, and defined timeframes vary from several hours to longer-term. In a recent review by Bae, 48 hours post-operatively was suggested as a reasonable timeframe; however, in practical terms, a palsy that occurs following normal facial nerve function has been witnessed post-operatively can be considered delayed.Reference Bae, Kwak, Nam and Jung1
Aetiology of delayed facial palsy
The aetiology of DFP is contested, and eliciting the exact cause of DFP following otological surgery is challenging. The most common theories are of viral reactivation of a latent virus, post-surgical oedema of the nerve or delayed insult to a dehiscent nerve. Other causes, such as retrograde injury to the nerve following transection of the chorda tympani, have also been proposed.
Viral reactivation
The most commonly proposed mechanism for DFP is the reactivation of a latent herpes virus at the geniculate ganglion secondary to surgical stressors on the immune system.Reference Guthikonda, Pensak and Theodosopoulos58 Elevated serum antibody titres of HSV type 1 are prevalent in the general population. They are found asymptomatically in 70 per cent of 25-year-olds and more than 90 per cent of 60-year-olds.Reference Guthikonda, Pensak and Theodosopoulos58
The exact mechanism of reactivation is unclear. PriceReference Price59 postulates that DFP is a result of the insult of the facial nerve at the peri-geniculate region secondary to skin stimuli such as ultraviolet light or systemic stimuli such as immune stressors. In one study, elevated HSV titres in patients with DFP compared to matched controls suggested that viral reactivation with HSV has a role to play.Reference Salvinelli, Casale, Vitaliana, Greco, Dianzani and D’Ascanio44 Furthermore, MRI studies on patients with DFP identified enhancement of the labyrinthine portion of the facial nerve with extension towards the fundus of the internal auditory canal, similar to that witnessed in viral reactivation, suggesting a potentially similar pathogenesis.Reference Cohen, Balaker, Kirsch, Mendelsohn and Ishiyama19
We found that only a minority of authors tested viral antibodies, and of those tested, only a minority were proven to have active infection, as indicated by IgM antibody titres (one of 15 cases for HSV and 10 of 21 cases for VZV). Previous infection was more prevalent, as suggested by raised IgG titres, was identified in 14/ of 15 of those tested for HSV IgG and in 18 of 20 of those tested for VZV IgG. Despite this, typical vesicles associated with VZV reactivation were only reported in one case by Di Berardino and Zanetti, suggesting most acute infections occur without vesicles.Reference Di Berardino and Zanetti20
Thermal injury oedema to the FN
Multiple authors attributed the proximity of the drill burr or KTP laser to the FN canal, resulting in heat propagating into the FN canal and resultant neuropraxia of the nerve to the aetiology of DFP.Reference Black, Bailey, Albert, Leighton, Hartley and Chatrath17, Reference Eskander, Holler and Papsin22–Reference Fayad, Wanna, Micheletto and Parisier24, Reference Mills, Szymanski and Abel38–Reference Ng and Maceri40, Reference Stratigouleas, Perry, King and Syms49 Mills et al.Reference Mills, Szymanski and Abel38 performed an in vitro study of FN canal temperature and found a temperature of 39 degrees for microdrill procedures and 52.2 degrees for laser procedures, suggesting both instruments are a source of heat and a potential insult to the nerve. Xu and Pollock studied the effects of heating a rodent sciatic nerve to 47 degrees for two 10-minute periods and found an immediate fall in blood flow, peri-vascular oedema and degeneration of the myelinated nerve fibres.Reference Xu and Pollock60 What temperature the FN can tolerate and the duration it can tolerate this for before undergoing injury is unknown.
Another cause of DFP postulated by Althaus is post-surgical oedema secondary to retrograde oedema from a directly traumatised, stretched or severed chorda tympani. This theory, however, is not widely accepted.Reference Althaus and House13
FN canal dehiscence
The third proposed theory for DFP is a delayed insult to a dehiscent FN. Moreano et al. examined 1000 temporal bones and discovered that 56 fallopian canals were dehiscent.Reference Moreano, Paparella, Zelterman and Goycoolea61 Similar rates were echoed by Baxter, who reported a 55 per cent rate of dehiscence in their review of 535 temporal bones.Reference Baxter62 The most common site of dehiscence is the region adjacent to the oval window, which was seen in 73.5 per cent of cases.Reference Moreano, Paparella, Zelterman and Goycoolea61, Reference Baxter62 From the articles included within this review, fallopian canal dehiscence was noted in five studies.Reference Lovin, Sweeney, Chapel, Alfonso, Govil and Liu6, Reference Kamalden, Yusof and Misron31, Reference Shea and Ge46, Reference Thorne, Dunham and Tom52, Reference Xu, Liu, Zuo, Wang and Wang55 Xu et al. found a significant association between those with DFP and fallopian canal dehiscence (p < 0.01).Reference Xu, Liu, Zuo, Wang and Wang55 Li et al.Reference Li, Chan, Chen, Kuo, Wang and Chang33 specifically reported using cotton material for the protection of the middle ear during surgical sterilisation to avoid irritation of the facial nerve. Kamalden et al.Reference Kamalden, Yusof and Misron31 attributed an ear pack compressing a dehiscent FN as the cause for DFP in their case.
FN monitoring
FN monitoring was employed in 12 studies, and in all cases, integrity testing was intact at the end of the procedure, suggesting that the nerve was intact and functioning intact at the time of surgery. This is expected as delays occur following a period of normal facial nerve function.
Management and therapy
Optimum treatment strategies for DFP remain inconclusive, mainly as a result of a lack of randomised controlled studies for comparison. Despite this, the prognosis of those with DFP remains excellent. Recovery data were available for 174/201 patients, with a full recovery achieved in 93 per cent of those with DFP following otological surgery.
In most studies, corticosteroids as monotherapy (n = 59) or as combination therapy (n = 50) were the modality of choice, largely based on strong evidence for corticosteroids in Bell’s palsy where a viral reactivation is the proposed causation and based on the role of corticosteroids in reducing oedema, inflammation in addition to its regenerative effects after peripheral nerve injury.Reference Couch, Hayward, Baum, Sakthiyendran, Harder and Hernandez63 Two authors reported on surgical decompression of the facial nerve following of DFP.Reference Durcan, Shea and Sleeckx21, Reference Gyo and Honda25
The optimum regimen, whether it be corticosteroid monotherapy or in combination with anti-viral therapy, is debatable. In a Cochrane review by Gagyor et al.Reference Gagyor, Madhok, Daly and Sullivan64 the use of anti-virals for Bell’s palsy was reviewed across 14 studies and 2488 participants. They found that a combination of anti-virals and corticosteroids may have little or no effect on rates of incomplete recovery in people with Bell’s palsy compared to corticosteroids alone (RR 0.81; 95% confidence interval [CI] 0.38–1.74). Furthermore, recovery rates were better in participants receiving corticosteroids alone than anti-virals alone (relative risk [RR] 2.69; 95% CI 0.73–10.01). The rate of incomplete recovery was lower with anti-virals and corticosteroids than with placebo or no treatment (RR 0.56; 95% CI 0.42–0.76). Anti-virals alone had no clear effect on incomplete recovery rates compared with placebo (RR 1.10; 95% CI 0.87–1.40). For people with severe Bell’s palsy (HB grade 5+), the combination of anti-virals and corticosteroids had no clear effect on incomplete recovery at month 6 compared to corticosteroids alone (RR 0.82; 95% CI 0.57–1.17).
In this review, 59 patients were treated with corticosteroid monotherapy, of which 56 improved to normal function. ThreeReference Jeppesen and Faber29, Reference Ng and Maceri40 did not return to normal function. Insufficient data were provided by the authors of these studies to elicit any comparisons between these patients who did not recover to full function. Seventeen were treated with corticosteroid and anti-viral therapy in the post-operative period, of which 12 returned to normal function and two regained an unspecified level of function. Of the three patients reported by Basavaraj who did not regain full function, no explanation is offered as to why this may be the case.
Despite the majority of studies employing corticosteroids as monotherapy or in combination with other therapies, several studies contested the benefit of corticosteroids in DFP. Marioni et al.Reference Marioni, de Filippis, Gaio and Staffieri34 reviewed the available literature in 2002 and concluded that corticosteroid therapy does not accelerate the improvement of DFP after stapedectomy. They conducted a Student’s t-test on unpaired samples from their review of the literature and found no statistical difference between patients treated with (n = 9) and without (n = 6) corticosteroid therapy. This sentiment that corticosteroids conferred no clear advantage in recovery time was echoed by Smith et al. (n = 6) who treated patients with a DFP either with dexamethasone or observation. The mean duration of palsy in the dexamethasone group was 16 days (however, when adjusted for a patient whose palsy lasted for 34 days and was associated an aberrant nerve, the mean duration was seven days) versus eight in the untreated group.Reference Smith, Simon and Ramalingam48
In two studies, pre-operative infection with the herpes viruses was taken into consideration. Salvinelli et al.Reference Salvinelli, Casale, Vitaliana, Greco, Dianzani and D’Ascanio44 treated all patients with a history of herpes labialis with pre-operative acyclovir and demonstrated a significant reduction in the prevalence of post-stapedectomy DFP (7/114 vs 0/35; p < 0.001). This author advocates that all patients undergoing otological surgery with previous evidence of herpes infection being treated with pre-operative anti-viral therapy. Bae et al. investigated the post-operative use of conservative treatment, corticosteroid therapy, anti-viral therapy and a combination of corticosteroid and anti-viral therapy for DFP following otological surgery, with further stratification in individuals which had previous facial palsy, herpes labialis or raised viral titres. They did not report a significant difference in complete recovery rate (p = 0.285) and period of paralysis (p = 0.116) between the groups, suggesting post-operative anti-viral therapy, even with a history of prior herpes infection, has no additional effect to observation or in combination with steroids.Reference Bae, Kwak, Nam and Jung1
The role of primary decompression in viral reactivation is of debate, but currently there is little evidence for it. In this review, two authors employed facial nerve decompression; Durcan et al. in 1967 in the primary setting for decompression of a “red and bulging” facial nerve overlying the round window nice following revision stapes procedure (HB grade unknown) and more recently by Gyo and Honda in a case of DFP with increased VZV titres but with no improvement in the dense (HB grade VI) palsy despite corticosteroid therapy.Reference Durcan, Shea and Sleeckx21, Reference Gyo and Honda25 In an attempt to achieve some facial nerve function in patients with a dense grade 6 palsy, with failure to improve with corticosteroid therapy and that demonstrates prognostically poor outcomes indicated by nerve conduction studies, there may be a role for facial nerve decompression. This must however, be balanced with the largely unknown outcomes and the potential significant morbidity of a middle cranial fossa approach that would be required to access the peri-geniculate region.
The main limitation of this narrative synthesis is the inability to perform a meta-analysis due to the inconsistency of reported outcomes.
Conclusion
DFP is a rare condition with an excellent prognosis, as most cases resolve completely. The leading theory regarding its cause involves viral reactivation within the geniculate ganglion. However, delayed facial nerve injury due to neuronal oedema triggered by stimuli such as heat may also contribute to its aetiology. Corticosteroids remain the cornerstone of treatment. The use of anti-viral therapy in DFP may be beneficial in the pre-operative setting for patients with a history of herpes infection. However, evidence from studies on Bell’s palsy and findings from this review do not support its routine use post-operatively. Surgical intervention is not indicated as a primary treatment for DFP, but it may be considered in cases of severe paralysis that fail to respond to medical therapy and show signs of a poor prognosis.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022215125103435.
Acknowledgements
Thanks to Matthew Stone, Clinical Effectiveness Librarian, Health Library, Royal Stoke University Hospital, ST4 6QG for assistance and for independently verifying the literature search.
Financial support
Nil.
Competing interests
Nil to declare.
