Efficacy of Additional (Selective) Vidian Neurectomy in Treating Chronic Rhinosinusitis with Nasal Polyp: A Systematic Review and Meta-Analysis Open Access

Managing chronic rhinosinusitis with nasal polyposis (CRSwNP) poses persistent challenges despite medical and surgical interventions. The chronic nature of the condition and its refractory course significantly impact patients’ quality of life [1, 2]. Also, patients diagnosed with both CRSwNP and allergic rhinitis (AR) exhibit heightened clinical manifestations and greater disease severity than do cohorts presenting with AR or CRSwNP alone [3]. Primary therapeutic intervention typically entails medical management involving topical corticosteroids and nasal saline irrigation [1, 2]. Surgical intervention is necessary for CRSwNP patients who do not respond satisfactorily to conventional medical therapy. The conventional treatment approach for individuals with concurrent CRSwNP and AR has involved surgical intervention followed by adjunctive therapy with nasal corticosteroid sprays or other anti-allergic agents for AR control. However, recurrence rates of CRSwNP postsurgical intervention range from 35% at 6 months to 38% at 12 months postoperatively, with frequent symptom persistence [4]. Autonomic dysfunction mediated by the vidian nerve and posterior nasal nerve often underlies persistent symptoms in such patients [5, 6]. Although vidian neurectomy has conventionally been employed in endoscopic sinus surgery (ESS) to alleviate symptoms in CRSwNP patients, it can lead to complications such as dry eyes and keratitis by disrupting secretomotor fibers innervating the lacrimal gland, nasal mucosa, palatine mucosa, and pharyngeal mucosa [7, 8]. Alternatively, selective transection of the posterior nasal nerve and/or pharyngeal nerve resection, which innervates secretomotor fibers supplying the nasal mucosa, offers a potential solution to alleviate rhinorrhea, sneezing, and itching via a sensory blockade. This selective vidian neurectomy (SVN) approach presents advantages over vidian neurectomy by mitigating risks of ocular dryness and palatal numbness [9]. Although the efficacy of vidian neurectomy and SVN has been established in patients with AR, its role in the management of CRSwNP remains uncertain. The present study aims to assess the effect of ESS with additional (selective) vidian neurectomy in patients with CRSwNP compared to that of conventional ESS. We also investigated whether there was a difference in the effectiveness of conventional vidian neurectomy and SVN.

This research was conducted in accordance with the principles established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The PRISMA Checklist is provided in Supplementary Table 1 (for all online suppl. material, see https://doi.org/10.1159/000546116).

Studies were collected from PubMed, SCOPUS, Embase, Web of Science, Google Scholar, and the Cochrane database up to Oct 2024. Key search terms and keywords included “vidian neurectomy,” “selective vidian neurectomy,” “posterior nasal neurectomy,” “sinus,” “paranasal sinuses,” “chronic sinusitis,” “paranasal sinusitis,” and “nasal polyp.” Online Supplementary Table 2 shows the detailed search strategy. The present investigation adhered to the PICO (population, intervention, comparison, and outcome) framework in the following manner: population: patients with CRSwNP; intervention: ESS with vidian neurectomy or SVN (posterior nerve resection and/or pharyngeal nerve resection); comparison: ESS only; and outcome: Lund-Kennedy (LK) score scale, patient-reported nasal score (visual analog scale; VAS), and patient-reported quality of life score (Rhinosinusitis Disability Index [RSDI] scoring for quality of life or Rhinoconjunctivitis Quality of Life Questionnaire [RQLQ]); and incidence of recurrence of CRS and adverse effects including eye symptoms and epistaxis.

Two independent literature reviewers (D.H.K. and S.H.H.) reviewed and screened the titles and abstracts for all searched studies and excluded studies not related to the subject. If the abstract alone could not help determine whether or not to include it, the full text was assessed. If opinions differed on inclusion of any study, a third reviewer (D.W.J.) was consulted. Studies were excluded if they addressed chronic sinusitis without nasal polyp or with concurrent other tumorous conditions.

Figure 1 depicts a flowchart delineating the process for selecting studies for analysis. The study protocol was duly registered on the Open Science Framework (https://osf.io/rq8kp).

Data extraction from eligible studies included information on patient demographics, the specific scale and scoring systems used for assessing endoscopic findings, patient-reported nasal symptom scores, quality of life scores, and p values to compare treatment outcomes between the intervention group (ESS plus neurectomy) and the control group (ESS only). This information was systematically organized according to a standardized format [10‒12]. Subsequent to extraction, a comparative analysis between the treatment and control groups was performed. Assessment of bias risk in non-randomized controlled studies employed the Newcastle-Ottawa Scale, whereas randomized controlled trials were evaluated using the Cochrane risk of bias tool.

Meta-analyses were performed using R statistical software version 4.4.0 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables from the original data were analyzed using the standard mean difference (SMD), and rate-related outcomes were evaluated based on odds ratio (OR). Heterogeneity was assessed using the I² test, which quantifies the extent of variation across studies attributable to heterogeneity rather than chance, ranging from 0 (indicating no heterogeneity) to 100 (suggesting maximum heterogeneity). All findings are presented with 95% confidence intervals (95% CI), and p values were calculated as two-tailed. In instances of significant heterogeneity among outcomes (defined as I² >50), the random-effects model, following the DerSimonian-Laird approach, was applied. This model acknowledges potential differences in treatment effects across studies, assuming a normal distribution of the effects. Conversely, outcomes lacking significant heterogeneity (I² <50) were analyzed using the fixed-effects model, which assumes a common population for all studies and employs the inverse variance method. Additionally, sensitivity analyses were conducted to evaluate the influence of each individual study on the overall meta-analysis results.

Eight articles involving a total of 565 participants were incorporated into the analysis [13‒20]. Detailed study attributes are delineated in Table 1, while the outcomes of bias assessment are elucidated in online supplementary Table 3 and 4. Due to the limited number of studies included for each outcome, falling short of the requisite threshold (<10) for robust evaluation of funnel plot asymmetry or advanced regression-based assessments, publication bias was not assessed.

The additional (selective) vidian neurectomy group showed a significantly improved LK score (SMD = 0.6014 [0.3777; 0.8251]; I² = 4.9%) and patient-reported quality of life score (SMD = 0.4476 [0.0208; 0.8744]; I² = 58.5%). This treatment also improved patient-reported nasal symptom-related VAS (SMD = 0.8946 [0.6211; 1.1681]; I² = 74.0%) (Fig. 2).

Significant inter-study heterogeneity (I² >50%) was found for the patient-reported quality of life score and VAS. The overall analysis did not consider the neurectomy extent (vidian neurectomy or SVN), which could have contributed to the considerable heterogeneity (exceeding 50%) observed across the results reported in all studies. To solve this, we performed subgroup analyses, in which both types of vidian neurectomy resulted in a change in LK score from baseline (SVN; SMD = 0.5503 [0.2870; 0.8137]; I² = 20.0%/vidian neurectomy: SMD = 0.5984 [0.0068; 1.1900]; I² = 59.0%; p = 0.8843). Conventional vidian neurectomy showed better improvement for patient-reported nasal symptom-related VAS (SVN: SMD = 0.7142 [0.4634; 0.9650]; I² = 52.8%/vidian neurectomy: SMD = 1.3773 [0.9192; 1.8355]; I² = 72.7%; p = 0.0128), but there was no significant difference in the changes of patient-reported quality of life scores from baseline between the two groups (SVN; SMD = 0.4773 [0.0638; 0.8908]; I² = 48.1%/vidian neurectomy: SMD = 0.5249 [−0.9557; 2.0055]; I² = 81.4%; p = 0.9516).

Additional (selective) vidian neurectomy did not significantly induce postoperative bleeding (OR = 0.9920 [0.3961; 2.4842]; I² = 0.0%) or dry eye-related discomfort (OR = 65.3560 [0.1044; 40,908.1619]; I² = 88.9%) compared to the control group (Fig. 3a, 3b and 3c). However, there was no significant difference in the recurrent rate of CRSwNP during the follow-up periods (OR = 0.7563 [0.3928; 1.4563], I² = 13.2%) between the two groups. Also, in the subgroup analyses of the neurectomy extent (conventional vidian neurectomy vs. SVN), conventional vidian neurectomy (OR = 1,727.2857 [87.1494; 34,234.4842]) led to significantly more frequent dry eye than did SVN (OR = 2.4194 [0.1114; 52.5255]) (p = 0.0027). However, there was no statistically significant difference in postoperative bleeding (SVN: OR = 0.9806 [0.2348; 4.0960]; I² = 0.0%/vidian neurectomy: OR = 1.0000 [0.3019; 3.3120]; p = 0.9836).

For eosinophilic vs. non-eosinophilic CRSwNP analysis, there are two papers on HPF analysis. In patient-reported nasal symptom-related VAS, only the eosinophilic group (SMD = 0.6524 [0.3066; 0.9983] was included and significantly improved (Fig. 4a). The Lund-Kennedy score showed a greater difference in the eosinophilic group (SMD = 0.9128 [0.4401; 1.3856] vs 0.0000 [−0.9564; 0.9564]), but this was not statistically significant (p = 0.0935)(Fig. 4b). On the other hand, the patient-reported quality of life score was significantly improved in the non-eosinophilic group compared to the eosinophilic group (SMD = 1.3349 [0.2742; 2.3956] vs SMD = −0.1796 [−0.8946; 0.5354]), p = 0.0203)(Fig. 4c). However, because these comparisons are based on only two studies, comparative advantages should be judged carefully.

Regarding AR, there were no reports analyzing the results according to the presence or absence of AR. Except for Mittal et al. and Chen et al. (all included allergic and nonallergic patients), all other studies were conducted on AR. Therefore, we performed a comparison with papers targeting only AR patients, excluding reports by Mittal et al. and Chen et al. In reports that included only AR, the change in LK score was relatively small (SMD = 0.5507 [0.2891; 0.8123]; I² = 30.7% vs. SMD = 0.6014 [0.3777; 0.8251]; I² = 4.9%), but patient-reported quality of life score (SMD = 0.5312 [−0.0120; 1.0744]; I² = 64.0% vs. SMD = 0.4476 [0.0208; 0.8744]; I² = 58.5%) and nasal symptoms VAS (SMD = 0.9591 [0.6366; 1.2816]; I² = 77.4%) vs. 0.8946 [0.6211; 1.1681]; I² = 74.0%) had a relatively greater difference in change.

Sensitivity analyses were conducted through systematic evaluation of variations in pooled estimates, whereby the meta-analysis was iteratively performed excluding one study at a time. Notably, all outcomes remained congruent with the results above.

In this study, additional (selective) vidian neurectomy significantly improved the Lund-Kennedy score and sinusitis-related life quality scores compared to the control group. In subgroup analysis, conventional vidian neurectomy showed greater improvement of patient-reported nasal symptom-related VAS than SVN, but there was no significant difference in the changes of sinusitis-related life quality scores between the two groups.

In an analysis including both conventional vidian neurectomy and SVN, the adverse effect did not show a significant difference compared to the ESS-only group. However, subgroup analysis revealed that conventional vidian neurectomy led to significantly more frequent dry eye than did SVN. In principle, the SVN procedure does not involve the sacrifice of the lacrimal nerve, and therefore, dry eye should not occur. However, one report indicated that while no actual dry eye was observed [13], patients experienced discomfort due to a relatively reduced sensation of tear production. For the purpose of strict adverse effect enrollment, this was included under the category of dry eye.

In patients with CRSwNP, the predominant Th2 inflammatory response often leads to increased eosinophil levels and elevated IgE, exacerbating nasal polyp symptoms. Mechanistically, nasal cavity inflammation disrupts nasal mucosal integrity, impairs ciliary function, and exacerbates nasal polyp symptoms, indicating an interdependent relationship between inflammation and symptom severity [21, 22]. Zhou et al. [18] analyzed eosinophilic and non-eosinophilic CRSwNP separately, and found that SVN reduces the recurrence rate in patients with eosinophilic CRSwNP. However, Lund-Kennedy score and the patient-reported scores showed controversial results. Also, although there were no reports comparing AR and non-AR patients, the results of reports that included only AR patients showed relatively low LK scores and relatively high patient-reported quality of life scores and nasal symptoms VAS. Therefore, the effect of neurectomy on eosinophilic CRSwNP or AR would require further study.

Histopathologically, nasal polyps manifest as mucosal edema characterized by basement membrane thickening, epithelial layer proliferation, glandular hyperplasia, edema, fibrosis, and stromal cellular infiltration [6]. (Selective) vidian neurectomy demonstrates notable efficacy, particularly in managing severe rhinorrhea because it effectively disrupts parasympathetic nerve fibers, leading to a significant reduction in nasal secretions [6]. The procedure’s effectiveness is attributed to decreased mucous gland secretions and basement membrane thickening [23, 24]. Moreover, (selective) vidian neurectomy can impede the action of neuropeptides, such as substance P, calcitonin gene-related peptide, and vasoactive intestinal peptide, released from non-adrenergic noncholinergic nerves in the nasal mucosa, which are implicated in nasal polyp formation and growth [25, 26]. Blocking these nerves may disrupt the pathway through which eosinophil-generated mediators activate sensory nerves [14, 25]. These findings are evidence of additional benefits of combined SVN and ESS in patients with CRSwNP. However, although heterogeneous results were reported across articles, there was no significant difference in recurrence rates between the two groups in the meta-analysis. This result suggests that additional (selective) vidian neurectomy is helpful in alleviating postoperative symptoms but has limitations in changing the clinical course of the disease. In addition, the effect of neurectomy is expected to be primarily effective in treating vasomotor symptoms such as sneezing and watery rhinorrhea, but most studies only targeted patients with AR and the difference between AR vs. non-AR was limited. There were no comparative papers on non-AR. Therefore, further studies on the presence of AR are needed.

There were differences between studies regarding improvement after (selective) vidian neurectomy with ESS. Mittal et al. [19] reported that SVN had an additional effect on symptom improvement up to 4 weeks after surgery, but there was no significant difference from the ESS-only group with additional time. Meanwhile, articles that reported limited inclusion criteria to patients with AR reported longer sustained effects. Gad et al. [17] reported that there was no significant difference in the symptoms of the two groups immediately after surgery, but as several months passed, additional improvement in symptoms was observed in the (selective) vidian neurectomy group. Zhou et al. [18] and Qi et al. [15] reported that the (selective) vidian neurectomy group showed significant improvement in symptoms compared to the group that continued to perform only ESS at the 2-year follow-up. Therefore, if careful patient selection is performed, long-term effects can be expected. There have been previous reports that analyzed CRSwNP and vidian neurectomy [27, 28]. However, these studies were either limited to AR patients or did not include subgroup analyses of the differences in surgical effects between ECRS and non-ECRS, nor did they address the differences in effects between conventional vidian neurectomy and SVN. As a result, these studies were unable to provide clear recommendations regarding which surgical technique is most effective for patients with different types of CRS (ECRS or non-ECRS). Therefore, we included more recent studies, aiming to provide more specific and useful information for clinicians.

Although this meta-analysis has the strength of confirming the effects and side effects of SVN in addition to ESS surgery, it has several limitations. First, there may be differences in outcomes due to differences in patient demographics in the included studies such as allergy, asthma, previous surgery, type and extent of surgeries, and postoperative treatment. In the past, vidian neurectomy was frequently performed using a physical method. This approach involved making an incision with an elevator or sickle knife at the posterior end of the middle turbinate (sphenopalatine foramen), followed by mucoperiosteal dissection. Once the neurovascular bundle was exposed, it was transected and coagulated. In recent years, however, ablation techniques have become the preferred method for vidian neurectomy. In most cases of SVN, radiofrequency ablation is primarily utilized. Posterior nasal nerve ablation is typically performed by ablating the area around the posterior attachment of the middle turbinate. For pharyngeal neurectomy, the sphenoid process of the palatine bone is removed from the posterior portion of the sphenopalatine foramen to expose the neurovascular bundle within the palatovaginal canal, after which the pharyngeal nerve is completely ablated. These factors may affect results. A meta-analysis was possible for eosinophilic vs. non-eosinophilic CRSwNP, but since only two study was analyzed, the results should be interpreted with caution. Also, most of the reports limited the inclusion criteria to patients with AR. Therefore, comparative analysis of the differences in effectiveness between AR and non-AR patients would also be necessary. However, since the treatment effect was compared by making the control group and treatment group similar during the comparison, heterogeneity is not expected to have a significant impact on the analysis of the effect depending on treatment status. In addition, the majority of reports are from China, and the rest are from India and Egypt, presenting a lack of diversity among countries. Second, there is a lack of well-designed studies such as double-blind, randomized, placebo-controlled trials. There is a need to secure additional evidence to confirm the heterogeneous clinical course and long-term effects of each report to date.

Selective or conventional vidian neurectomy with ESS demonstrates superior efficacy in managing patients with CRSwNP by ameliorating both endoscopic findings and associated symptoms compared to those of ESS alone. Although conventional vidian nerve resection shows improved efficacy in relieving nasal symptoms compared to SVN, and the improvement in sinus-related symptoms is similar, there are concerns about side effects. Therefore, clinically, it would be more appropriate to first consider additional SVN if necessary along with ESS in patients with CRSwNP.

Our Institutional Review Board (Institutional Review Board of the Catholic University of Korea) approval was waived for this study because it is a systematic review and meta-analysis that based exclusively on published literature. Informed consents were not required because this study is based exclusively on published literature.

The authors declare that there are no competing interests.

This study was not supported by any sponsor or funder.

D.H.K.: study conception and design, acquisition of data, analysis and interpretation of data, drafting the article and revisions, final approval of article; D.W.J.: analysis and interpretation of data, drafting the article and revisions, final approval of article; S.H.H.: study conception and design, acquisition of data, analysis and interpretation of data, drafting the article and revisions, and final approval of article.

The raw data of individual articles used in this meta-analysis are included in the main text or supplementary data. Further inquiries can be directed to the corresponding author.