Introduction
Chronic rhinosinusitis with nasal polyps is an inflammatory disease with a high risk of recurrence, and patients often require repeated surgery or recurrent or continuous medical treatment.Reference Wu, Ting, Platt, Tierney and Metson1 Although the recurrence rate of chronic rhinosinusitis with nasal polyps has been reported as being as high as 90 per cent in different studies after functional endoscopic sinus surgery (FESS), this rate has been calculated as approximately 15 per cent in systematic reviews.Reference Wu, Ting, Platt, Tierney and Metson1–Reference Prasad, Fong and Ooi3 Despite all the technical advances (navigation systems, three-dimensional visualisation) in FESS, repeat operations continue to be required in some cases.Reference Alsaleh, Manji and Javer4,Reference Philpott, Hopkins, Erskine, Kumar, Robertson and Farboud5 Likewise, although biologic agents (e.g. dupilumab) in medical treatment are approved for the management of chronic rhinosinusitis with nasal polyps, they are not cost-effective against FESS.Reference Bachert, Zhang, Cavaliere, Weiping, Gevaert and Krysko6,Reference Scangas, Wu, Ting, Metson, Walgama and Shrime7
Risk factors for recurrence of nasal polyps after FESS have been discussed in previous literature, but probably the most important cause of failure is related to the inflammatory processes of the disease endotypes.Reference Loftus, Soler, Koochakzadeh, Desiato, Yoo and Nguyen8,Reference Patel, Kern, Bernstein, Hae-Sim and Peters9 Type 2 inflammation (higher expression of interleukin-5, immunoglobulin E and eosinophil cationic protein) was proven to be a strong predictor of recurrent disease in the majority of cases with eosinophil clusters.Reference Jonstam, Swanson, Mannent, Cardell, Tian and Wang10,Reference Wei, Liu, Zhang, Liu, Du and Liu11 Frequently studied clinical factors for revision FESS include: tissue eosinophilia, high blood eosinophil or basophil count, allergic rhinitis, asthma, aspirin sensitivity, sinus involvement, paranasal structure, radiological findings (osteitis), previous number of FESS procedures, type of rhinosinusitis, duration of medication use, and type of medication used.Reference Stein, Jafari and DeConde2,Reference Loftus, Soler, Koochakzadeh, Desiato, Yoo and Nguyen8,Reference Musy and Kountakis12–Reference Rajwani, Manji, Finkelstein-Kulka, Habib, Alsaleh and Macias-Valle15
Koskinen et al. demonstrated significant predictive risk factors for revision FESS during a five-year follow up; these were: allergic rhinitis, aspirin-exacerbated respiratory disease, recurrent polyps, need for regular intranasal corticosteroid treatment, pre-operative per oral corticosteroid use, previous sinus surgery or polypectomy, and current ethmoidectomy.Reference Koskinen, Salo, Huhtala, Myller, Rautiainen and Kääriäinen16 Stein et al. reported the independent risk factors for revision FESS as being female gender (adjusted odds ratio = 1.20), public insurance (adjusted odds ratio = 1.10) and Hispanic ethnicity (odds ratio = 0.86) in a large population-based review of all types of chronic rhinosinusitis.Reference Stein, Jafari and DeConde2 Wu et al. prospectively showed that nasal allergy (odds ratio = 9.2) and Lund–Mackay score (odds ratio = 1.29) were both independent risk factors for revision FESS in 84 patients with chronic rhinosinusitis with nasal polyps.Reference Wu, Lee, Huang, Chang and Fu14 In a retrospective analysis, Loftus et al. indicated independent risk factors associated with revision FESS as being: younger age (odds ratio = 1.1), prior surgery (odds ratio = 3.3), longer follow up (odds ratio = 1.1) and surgery conducted before 2009 (odds ratio = 2.4).Reference Loftus, Soler, Koochakzadeh, Desiato, Yoo and Nguyen8
Histopathological features and inflammation type provide more precise information to predict surgical failure in patients with chronic rhinosinusitis with nasal polyps. Unfortunately, if treatment with a biologic molecule is planned and the specific receptor is not targeted, it may not be cost-effective to determine the inflammatory profile of polyp tissue, and is not as common as FESS therapy in clinical practice. However, questioning the basic clinical features and evaluation of paranasal sinus computed tomography (CT) findings (phenotype) among patients who will undergo FESS may help to predict surgical failure simply and accurately.
Materials and methods
In this retrospective study, the clinical records of patients with diagnosed chronic rhinosinusitis with or without nasal polyps were analysed. The study was conducted between 2011 and 2018 in İzmir Atatürk Research and Education Hospital (affiliated hospital of Katip Çelebi University Medical School). The data of patients with chronic rhinosinusitis with nasal polyps who underwent FESS consecutively were detailed.
The operations were performed under general anaesthesia, with a 4 mm diameter, 170 mm length, 0- or 30-degree rigid sinuscope. No specially described or manufactured instruments or navigation systems were used. The diagnosis of chronic rhinosinusitis with nasal polyps and the decision to undertake revision FESS were made in line with the European Position Paper on Rhinosinusitis and Nasal Polyps (‘EPOS’) 2012.Reference Fokkens, Lund, Mullol, Bachert, Alobid and Baroody17
Patients older than 18 years who were operated on by surgeons with similar experience were included in the study. Patients with a diagnosis of neoplasm, antrochoanal polyp, granulomatous and autoimmune disease, and craniofacial anomalies were excluded, as were those lacking data.
Clinical records were reviewed to collect information on: demographics, patient history, smoking status, intranasal and systemic steroid use, asthma, diagnosis of allergic rhinitis, aspirin sensitivity, and radiological features (on paranasal sinus CT, 0.5 mm thick, high-resolution sections of all three anatomical planes; Toshiba Activion™ 64-multislice CT system). Patients who did not use regular nasal steroids for at least one month were included in the group of patients who did not use nasal steroids. The predominance of eosinophilic infiltration, presence of mucins, and concomitant fungal infection in pathological specimens were examined. In addition, the total follow-up period, the occurrence of revision FESS, the number of revision FESS procedures carried out, and the duration of all revision surgical treatment were determined. Lund–Mackay scores for both sides were calculated and summated to obtain a single score (minimum = 0, maximum = 24).Reference Lund and Mackay18 The Global Osteitis Scoring Scale score was calculated for 10 sinuses (right and left frontal, anterior ethmoid, posterior ethmoid, maxillary, and sphenoid; minimum = 0, maximum = 40).Reference Georgalas, Videler, Freling and Fokkens19
The patients were divided into two groups according to the occurrence of revision FESS. Study groups were compared for possible risk factors. Statistical analysis was performed using SPSS software, version 16 (IBM, Armonk, New York, USA). The statistical significance level was established at p < 0.05, with 95 per cent confidence intervals. Categorical data were analysed with a chi-square test. The data found to be statistically significant in the chi-square test were subsequently included in a binary logistic regression analysis, to identify the possible risk factors for revision FESS.
The authors assert that all procedures contributing to this work complied with the ethical standards of the relevant national and institutional guidelines on human experimentation, and with the Helsinki Declaration of 1975, as revised in 2008. The study was approved by the ethical committee of the affiliated university (approval number: 0292-24/06/2021).
Results
The mean age of the 212 patients included in the study was 42.4 ± 14.3 years. Of the 100 patients who underwent revision surgery, 72 patients (mean age, 40.1 ± 15.2 years) had only one revision procedure and 28 patients (mean age, 46.8 ± 15.5) had more than one revision procedure. Nasal steroids were used pre-operatively in 188 patients (88.6 per cent), and systemic steroids were used pre-operatively in 75 patients (35.3 per cent). Eosinophilia was detected in 62 (29.5–2 per cent) of the pathological specimens. Asthma was diagnosed in 48 patients (22.6 per cent), and aspirin sensitivity was reported in 18 patients (8.4 per cent). Forty-seven patients (22.7 per cent) used nasal steroids post-operatively. Most of the residual cells after initial FESS were observed in the frontal sinuses (n = 118; 59.0 per cent). A general summary of the findings is presented in Table 1.
Table 1. Overall summary of all findings
SD = standard deviation; pre-op = pre-operative; FESS = functional endoscopic sinus surgery; intra-op = intra-operative; post-op = post-operative
There were 112 patients (mean age, 42.8 ± 13.1 years) (85 males, 27 females) in the no-revision surgery group and 100 patients (mean age, 42.0 ± 15.5 years) (50 males, 50 females) in the revision surgery group (p < 0.001). The rate of pre-operative nasal steroid use was significantly higher in the revision surgery group, while pre-operative and intra-operative systemic steroid use was significantly higher in patients who did not undergo revision FESS (p < 0.001, 0.07 and < 0.001, respectively). In addition, the rates of asthma, aspirin sensitivity, smoking and a positive prick test result were significantly higher in the revision FESS group (p < 0.001, 0.001, < 0.001 and < 0.001, respectively). The distribution of other risk factors according to the occurrence of revision FESS is given in Table 2. There was no significant symptomatological difference between the study groups (Table 3). The mean Global Osteitis Scoring Scale score was 15.9 ± 9.1 in the revision FESS group and 15.8 ± 5.3 in the patients who did not undergo revision surgery. The mean Lund–Mackay score was 8.4 ± 2.5 in the revision FESS group and 8.2 ± 2.0 in the no-revision surgery group (Table 4).
Table 2. Chi-square statistics with frequencies of possible risk factors for revision FESS
* Indicates statistical significance (p < 0.05). FESS = functional endoscopic sinus surgery; pre-op = pre-operative; intra-op = intra-operative; post-op = post-operative
Table 3. Chi-square statistics with distribution of symptoms by groups with or without revision FESS
FESS = functional endoscopic sinus surgery
Table 4. Mean and standard deviation of scale variables for revision FESS
FESS = functional endoscopic sinus surgery; SD = standard deviation; max = maximum; min = minimum
Multivariate analysis (logistic regression) demonstrated that only gender, pre-operative nasal steroid use, pre-operative systemic steroid use, intra-operative systemic steroid use and smoking were significant risk factors for revision FESS (p = 0.034, 0.001, 0.010, 0.007 and 0.001, respectively) among those significant factors identified on univariate analyses (Table 5).
Table 5. Logistic regression of risk factors for FESS found to be significant on univariate analysis
FESS = functional endoscopic sinus surgery; Sig. = significance; OR = odds ratio; CI = confidence interval; pre-op = pre-operative; intra-op = intra-operative
Chi-square statistics with the frequency of risk factors for two or more multiple revision FESS procedures are given in Tables 6 and 7. According to the analysis performed after excluding patients who did not undergo revision FESS, only eosinophilia and aspirin sensitivity were found to be significant risk factors for multiple revision FESS procedures (p = 0.043 and 0.005, and odds ratio = 2.4 and 5.2, respectively). However, according to the multinomial regression analysis, only aspirin sensitivity was a significant risk factor (p = 0.013, odds ratio = 4.3).
Table 6. Frequencies of possible risk factors for revision FESS groups
Chi-square statistics were calculated following the exclusion of patients with no revision functional endoscopic sinus surgery (FESS). Pre-op = pre-operative; intra-op = intra-operative; post-op = post-operative
Table 7. Mean and standard deviations of scale variables
* P-values obtained from one-way analysis of variance comparisons. FESS = functional endoscopic sinus surgery; SD = standard deviation
The mean follow-up period for all patients was 52.1 ± 24.2 months. The mean follow-up period was 49.4 + 21.5 months for patients who did not undergo any revision surgery, 57.2 + 27.6 months for patients who underwent one revision FESS and 49.8 + 24.5 months for patients who had more than one revision surgery procedure. There was no significant difference between groups (p = 0.088, one-way analysis of variance).
Discussion
This study showed that the rates of asthma, aspirin sensitivity, smoking and a positive prick test result were significantly higher in the revision FESS group. Additionally, gender, pre-operative nasal steroid use, pre-operative systemic steroid use, intra-operative systemic steroid use and smoking were significant risk factors for revision FESS identified on univariate analysis and multivariate analysis. Eosinophilia and aspirin sensitivity were significant risk factors for multiple revision FESS procedures, but the multinomial regression indicated that only aspirin sensitivity was a significant risk factor.
According to the European Position Paper on Rhinosinusitis and Nasal Polyps 2020 classification, chronic rhinosinusitis with nasal polyps is defined as a subgroup of type 2 inflammation-associated (type 2 endotype) primary chronic rhinosinusitis.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma20 Aspirin-exacerbated respiratory disease continues to dominate as the main risk factor for FESS failure, which remains valid in different studies and meta-analyses. Aspirin-exacerbated respiratory disease or non-steroidal anti-inflammatory drug (NSAID) exacerbated respiratory disease is a chronic inflammatory disorder of the respiratory tract, with tissue eosinophilia occurring in patients with asthma and/or chronic rhinosinusitis with nasal polyps. Adelman et al. reported the rate of previous FESS as approximately 50–100 per cent (range, 1.9–5.4 operations) in their review of the management of NSAID-exacerbated respiratory disease (1618 studies with 686 patients).Reference Adelman, McLean, Shaigany and Krouse21 Aspirin-exacerbated respiratory disease or NSAID-exacerbated respiratory disease reflects a special subgroup of patients with chronic rhinosinusitis with nasal polyps; it has pathophysiology consistent with type 2 inflammation. In these patients, treatment with biologic therapies targeting cytokines (interleukins (IL)-4, IL-13, IL-5 and immunoglobulin E) may be considered, to avoid repeat surgical procedures or to be used in patients with a contraindication for FESS.Reference Damask, Ryan, Casale, Castro, Franzese and Lee22
High eosinophilic infiltration of the airway mucosa has been demonstrated regarding the pathophysiology of chronic rhinosinusitis with nasal polyps and asthma, and is a histochemical characteristic of type 2 inflammation. Epidemiological, pathophysiological and histological characteristics, and clinical and treatment outcomes of patients with asthma and chronic rhinosinusitis with nasal polyps, are significantly correlated, as defined in the concept of a ‘united airway disease’.Reference Yorgancioğlu, Kalayci, Kalyoncu, Khaltaev and Bousquet23,Reference Langdon and Mullol24 Furthermore, in cases of high eosinophilic infiltration (10 per high power field (400×) or higher) with clinically localised diffuse chronic rhinosinusitis, the clinical phenotypes have been defined as predominantly eosinophilic chronic rhinosinusitis and non-eosinophilic chronic rhinosinusitis.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma20 Existing literature seems to agree that eosinophilic infiltration of mucosa or polyp tissue significantly predicts revision surgery, polyp recurrence, and disease severity in the other subgroups of chronic rhinosinusitis (except eosinophilic chronic rhinosinusitis).Reference Nakayama, Yoshikawa, Asaka, Okushi, Matsuwaki and Otori25–Reference Virkkula, Penttilä, Vento, Myller, Koskinen and Hammarén-Malmi27 High rates of revision surgery in high tissue eosinophilia cases should never come as a surprise. In fact, treatment modalities that do not explain tissue eosinophilic infiltration and the immunohistochemical mechanism that leads to and targets these pathways are doomed to fail. As might be expected, revision FESS rates are higher in: patients who had their first operation at a younger age, in older patients, and in patients with long-term follow up.Reference Rajwani, Manji, Finkelstein-Kulka, Habib, Alsaleh and Macias-Valle15,Reference Loftus, Soler, Desiato, Koochakzadeh, Yoo and Storck28,Reference Bayer, Hamidovic, Besser, Mueller and Liu29
Essentially, FESS is recommended for the management of chronic rhinosinusitis, to enable the creation of a sinus cavity that: incorporates the natural ostium, allows adequate sinus ventilation, facilitates mucociliary clearance and facilitates the application of topical therapies.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma20 From the patient's perspective, FESS primarily aims to treat five key symptoms of: nasal obstruction, rhinorrhoea, postnasal drip, facial pain and anosmia.Reference Prasad, Fong and Ooi3 Nasal obstruction is the main symptom in patients with chronic rhinosinusitis with nasal polyps due to the mass of polyps.Reference Bachert, Bhattacharyya, Desrosiers and Khan30 Therefore, a functional FESS may be preferred to relieve nasal obstruction. However, it has been shown previously that a more radical FESS is required for olfactory dysfunction, for example. Thus, the concept of FESS failure will vary according to risk factors, co-morbidities, leading symptoms and the extent of FESS to be performed.Reference Benkhatar, Khettab, Sultanik, Laccourreye and Bonfils13,Reference Zhang, Zhang, Gao, Wang, Lou and Meng31
The socio-demographic characteristics of patients associated with the risk of revision FESS in the literature include: smoking, female gender, non-Hispanic origin and geographical location (Europe and Asia).Reference Stein, Jafari and DeConde2 Smith et al. reported a low risk rate for first revision FESS in males.Reference Smith, Orlandi, Oakley, Meeks, Curtin and Alt32 Wu et al. found that smoking decreased the disease-free period, and patients who smoked had a significantly shorter time to revision surgery.Reference Wu, Ting, Platt, Tierney and Metson1 In the most recent and comprehensive meta-analysis, only geographical location was discussed in addition to these factors, and other features were not mentioned.Reference Loftus, Soler, Koochakzadeh, Desiato, Yoo and Nguyen8 Although the risk attributed to socio-demographic characteristics has not been proven, the results in the literature and in our study are in agreement. Extended surgical procedures, radical FESS or Draf III procedures, have been associated with a reduced number of revision surgical procedures and a prolonged time to recurrence.Reference Koskinen, Salo, Huhtala, Myller, Rautiainen and Kääriäinen16,Reference Zhang, Zhang, Gao, Wang, Lou and Meng31,Reference Masterson, Tanweer, Bueser and Leong33 In addition, while the number of revision procedures increases in cases of simple polypectomies, the time to revision surgery is shortened. It has also been reported that adding middle turbinectomy to FESS additionally reduces the risk of revision in the long term.Reference Wu, Ting, Platt, Tierney and Metson1,Reference Musy and Kountakis12,Reference Benkhatar, Khettab, Sultanik, Laccourreye and Bonfils13 This study included patients with chronic rhinosinusitis with nasal polyps who underwent functional FESS, and the middle turbinate was not excised in any of the patients. None of the reported surgical-anatomical structures were identified as risk factors in this study.
In patients with chronic rhinosinusitis with nasal polyps, scores of the Lund–Mackay scale – which expresses the extent of the disease, and correlates with the treatment results and the severity of the disease – have previously been associated with the risk of revision FESS. Even significant cut-off scores of 4 or greater or 13.5 or greater have been reported.Reference Wu, Lee, Huang, Chang and Fu14,Reference Virkkula, Penttilä, Vento, Myller, Koskinen and Hammarén-Malmi27,Reference Younis and Ahmed34 Bone remodelling and osteitis correlate with disease severity in patients with chronic rhinosinusitis with nasal polyps, and reduce the chance of treatment success.Reference Benkhatar, Khettab, Sultanik, Laccourreye and Bonfils13,Reference Baban, Mirza and Castelnuovo35 Previously, Georgalas et al. showed a statistically linear relationship between revision surgery and osteitis scores.Reference Georgalas, Videler, Freling and Fokkens19 Few studies have used the Global Osteitis Scoring Scale as a risk factor for revision surgery. However, contrary to expectations and the literature, no significant relationship was found between the Global Osteitis Scoring Scale and revision surgery risk in this study.
Another factor influencing the risk of revision FESS is medical treatment regimens. Systemic versus nasal steroid applications show a significant difference in terms of the management of chronic rhinosinusitis with nasal polyps.Reference Loftus, Soler, Koochakzadeh, Desiato, Yoo and Nguyen8 The proportions of the patients using nasal steroids pre-operatively, systemic steroids pre-operatively and systemic steroids intra-operatively were identified as predictive risk factors for revision FESS in the presented study. Younis and Ahmed reported that post-operative courses of systemic or nasal steroids were not affected the revision FESS rates, in a similarly designed study.Reference Younis and Ahmed34 Likewise, we did not find a relationship between post-operative nasal steroid use and revision FESS. However, corticosteroids are the preferred group of medications for: treating severe disease, alleviating the pathology until the time of the operation, and suppressing recurrence after surgery. It can be predicted that treatment outcomes will be worse in patients who have to apply intensive corticosteroids. In fact, one of the goals of recurrent surgery is to protect the patient from the side effects of intensive steroid therapy.Reference Bachert, Bhattacharyya, Desrosiers and Khan30
The results of the present study are generally compatible with the literature. However, contrary to expectations, the Lund–Mackay and Global Osteitis Scoring Scale scores were not found to be associated with revision FESS. This may be because the sample is not homogeneous. Furthermore, it was not possible to retrospectively identify separate classes of rhinosinusitis, such as allergic fungal chronic rhinosinusitis and eosinophilic chronic rhinosinusitis, from the pathology reports. Prospective cohort studies will yield more valid results in this regard. This study also includes primary FESS surgical procedures performed in different institutions; thus, pre-operative care, medical therapies and minor interventions may have differed. Another major limitation is inadequacies in histopathological examination. Pathology reports collected retrospectively did not include information on evidence of type 2 inflammation, such as an eosinophil count or density.
• Chronic rhinosinusitis with nasal polyps has a high risk of recurrence, being predominantly type 2 inflammation mediated
• Functional endoscopic sinus surgery (FESS) has a major role in the disease management
• Aspirin-exacerbated respiratory disease, asthma and eosinophilic pattern are well-studied and confirmed risk factors for revision FESS
• Type 2 inflammation is a strong predictor of recurrence in most cases with eosinophil clusters
• Clinical and radiological factors (phenotypes) can be used to predict patients at high risk of recurrence
Conclusion
Chronic rhinosinusitis with nasal polyps has a significant risk for recurrence, due to the predominantly type 2 inflammation mediated nature of the disease. Functional endoscopic sinus surgery has, and will continue to have, a major role in the management of disease. Aspirin-exacerbated respiratory disease, asthma and eosinophilic pattern are well-studied and confirmed risk factors for revision FESS. Facilities for histopathological examination of the mucosa or polyp tissue are not widespread as to performing a functional FESS. It is important to identify those clinical and radiological factors (phenotypes) that can be used to make a valid prediction regarding patients at high risk for recurrence. Thus, these patients should be referred for consultation at appropriate institutions in order to receive the most suitable surgical method or treatment. This also allows more accurate decisions regarding the use and duration of pre- and post-operative medical adjuvant treatments.
Competing interests
None declared
